Resources Policy 44 (2015) 35–46
Contents lists available at ScienceDirect
Resources Policy
journal homepage: www.elsevier.com/locate/resourpol
Raw material criticality in the context of classical risk assessment
Simon Glöser a,n, Luis Tercero Espinoza a, Carsten Gandenberger a, Martin Faulstich b
a
Competence Center Sustainability, Business Unit Systemic Risks, Fraunhofer Institute for Systems and Innovation Research ISI, Breslauer Str. 48, 76139
Karlsruhe, Germany
b
CUTEC Institute, Chair of Environmental and Energy Technology, Clausthal University of Technology, Leibnizstraße 21 þ 23, 38678 Clausthal-Zellerfeld,
Germany
art ic l e i nf o
a b s t r a c t
Article history:
Received 18 September 2014
Received in revised form
23 December 2014
Accepted 31 December 2014
The rapid economic development of emerging countries in combination with an accelerating spread of
new technologies has led to a strongly increasing demand for industrial metals and minerals regarding
both the total material requirement and the diversity of elements used for the production of specific
high-tech applications. Several minor metal markets which are often characterized by high market
concentrations of raw material production at the country and the company level have shown high
turbulences since the beginning of the 21st century. This has led to growing concerns about the security
of raw material supply, particularly in established western economies. As a result, numerous studies on
supply risks and raw material criticality for different countries and regions were carried out recently. In
this paper, we discuss the methodology of raw material criticality assessment within a criticality matrix
which is a modification of a classical risk matrix. Therefore, we first provide an overview of the
approaches and results of major studies quantifying raw material criticality by means of a criticality
matrix. By applying a uniform scaling to the matrices of different recent studies, a direct comparison of
results and data interpretation was enabled. As shown in this paper, the close relation between the
criticality matrix and classical risk analysis within a risk matrix was overlooked in most studies which
may lead to misunderstanding and misinterpretation of the results. We posit that the interpretation of
the coordinates within the criticality matrix and the thresholds separating critical and non-critical raw
materials need to be revised by means of general risk definitions.
& 2015 Elsevier Ltd. All rights reserved.
Keywords:
Critical raw materials
Criticality assessment
Criticality matrix
Risk matrix
Supply risks
Introduction
A basic requirement for sustainable economic development and
the successful production of high technology applications is the secure
supply with raw materials, free from disruptions, disturbances and
bottlenecks leading to high commodity pricing and market volatility.
Most industrialized countries strongly depend on raw material
imports, as their domestic raw material deposits and exploitation activities are small (Behrens et al., 2007). Forced by the rapid growth of
emerging markets, particularly China, the increasing dynamics in the
development of new technologies and the increasing diversification of
metals with very specific properties needed for these high-tech
applications, the raw material supply situation has strongly deteriorated in the previous decade (Rosenau-Tornow et al., 2009).
Current supplies of raw materials are often characterized by high
concentrations of production both on the country and the company
level (Sievers and Tercero, 2012). This means mining activities are
n
Corresponding author. Tel.: þ 49 721 6809 387; fax: þ 49 721 6809 135.
E-mail address: simon.gloeser@isi.fraunhofer.de (S. Glöser).
URL: http://www.r-cubed-research.eu (S. Glöser).
http://dx.doi.org/10.1016/j.resourpol.2014.12.003
0301-4207/& 2015 Elsevier Ltd. All rights reserved.
limited to a few countries and basic raw material processing is
carried out by several large mining corporations with significant
power in oligopolistic markets. In this context, the distortion of
competition caused by export restrictions and the taxation of
specific high-tech metals in several emerging countries are a serious
threat to different industries as both higher prices and the limited
availability of essential raw materials compromise their competitiveness (e.g. Parthemore, 2011; Campbell, 2014; Massari and
Ruberti, 2013).
Furthermore, several mining countries are suffering from political instability and inadequate economic and social conditions. The
potential for political conflicts in these countries is high and poses a
latent threat to raw material supplies (Le Billon, 2001).
Regarding geology, different minor metals only occur within the
ores of major carrier metals and are therefore mainly jointly
extracted and refined as the separate production of these sideproduct or companion metals is usually economically not feasible
(Verhoef et al., 2004). This is of particular relevance if the demand
of the byproduct materials increases faster than the demand of the
main metal, for instance due to the non-substitutable use within an
emerging technology. In this case, supply cannot be independently
increased and will not be able to meet demand which can lead to
S. Glöser et al. / Resources Policy 44 (2015) 35–46
disproportionately high market prices and raw material scarcity
(Fizaine, 2013; Tercero, 2012).
Beside social, economic and political aspects of primary raw
material production, the ecologic implications of mining and material
processing are gaining increasing public attention which could also
affect supply security due to regulations and restrictions in mining
countries or potential environmental regulations and strong certification requirements in raw material consuming countries (Bleischwitz
et al., 2012; Norgate et al., 2007).
Finally, the market dynamics are increasingly forced by speculation on raw material pricing and raw material related commodity derivatives as financial markets and raw material markets
move closer together (Humphreys, 2010a; Tilton et al., 2011).
Hence, there is mounting anxiety that the development, the
commercialization and the use of new innovative technologies
might be negatively affected or prevented due to shortages and
high pricing in raw material markets (e.g. Hoenderdaal et al., 2013;
Novinsky et al., 2014). Therefore, the systematic evaluation of
supply risks, vulnerabilities and economic consequences of supply
restriction form scientific challenges at present. In this context, the
determination of raw material criticality is a key element in
quantifying and communicating economic vulnerabilities due to
insecure material supplies.
Supply risks or criticality may be assessed for an enterprise (e.g.
Duclos et al., 2008), a country (e.g. Erdmann et al., 2011), a region
(e.g. European Commission, 2010, 2014) or for the world (e.g.
Graedel and Nassar, 2013). Apart from the spatial dimension,
several studies particularly focus on raw material supply for
specific emerging technologies, with particular emphasis on
energy technologies (e.g. APS, 2011; Moss et al., 2011, 2013; U.S.
DoE, 2010, 2011). Based on the methods used, Erdmann and
Graedel (2011) classified major studies dealing with the quantification of raw material criticality and supply risks into three
categories:
1. Studies using the principle of a criticality matrix as a modification of a classical risk matrix in order to assess raw material
criticality.
2. Studies quantifying a single risk index which is calculated from
different sub indicators.
3. Studies working with scenario analysis and time series analysis
in order to forecast demand (and supply) side developments.
An overview of different publications based on the concepts described in items 1–3 is provided in the accompanying supplementary
information. In this paper, we focus on the concept of criticality
determination within a criticality matrix (item 1 above). While recent
publications in this field analyzed and discussed the choice and
weighting of underlying indicators used for the quantification of a
material's supply risk or its economic importance (Achzet and Helbig,
2013; Erdmann and Graedel, 2011), herein, we focus on the interpretation of a material's position within the criticality matrix in the
context of general risk analysis—the original inspiration for the
criticality matrix. As described in the following sections, the close
link between a classical risk matrix and the criticality matrix has
been at least partly overlooked in several previous studies. This can
potentially lead to misinterpretations which we intend to clarify in
this paper.
After a short review of the historic debate about critical raw
materials, we provide a detailed quantitative derivation of the
criticality matrix approach as a modification of a classical risk
matrix. Then we present the matrices of different criticality studies
from recent years and we compare the interpretation of the results
from different studies by projecting the materials' coordinates into a
matrix with uniform scaling of the axes and contour lines representing the criticality level.
Current and historic debate about critical raw materials
Despite the recently increasing interest in critical metals and
minerals, the topic of raw material supply security goes back to
early human civilizations (Buijs et al., 2012) and whole periods of
human history were named after the metals or alloys that dominated anthropogenic use like the “Copper Age”, the “Bronze Age” or
the “Iron Age” (NRC, 2008).
Regarding the 20th century, which is most relevant for current
supply aspects, the debate about the security of raw material supplies
was dominated by political conflicts such as the two World Wars or
the Cold War (Gandenberger et al., 2012). The term “critical raw
material” was first introduced in the “Strategic and Critical Materials
Stock Piling Act” from 1939 (Legislative Councel, 1939). The “President's Materials Policy Commission” was appointed by President
Truman in the early 1950s due to fears of raw material shortages not
only for the United States but for the whole western world (Mason,
1952). In the 1970s and 1980s, due to relatively high commodity
prices (Humphreys, 2010a), the two oil crises in 1973 and 1979
(Kesicki, 2010), the cobalt crisis in 1978 (Alonso et al., 2007) and not
least because of the Cold War, the awareness of import dependencies
and vulnerabilities was high (Humphreys, 2010b). This is evident
from several publications about strategic and critical raw materials
from that time (Haglund, 1984; Jacobson et al., 1988; Leamy, 1985;
Robinson, 1986) and official reports from governmental institutions
such as the U.S. Council on International Economic Policy (1974), the
Commission of the European Communities (1975) or the U.S.
Congressional Budget Office (1983). Furthermore, the issue of raw
material criticality was part of political initiatives like “The National
Critical Materials Act of 1984” (Committee on Science, 1984). However, after the collapse of the Soviet Union and due to continuously
decreasing commodity prices in the 1990s, the topic of non-fuel
minerals and metals supply security lost attention for more than a
decade (Humphreys, 1995).
This has substantially changed over the past years. Due to the
aforementioned current tensions in raw material markets, numerous studies about the quantification of supply risks of mineral and
metallic raw materials have been carried out in the past 10 years. A
literature review recently published by the UK Energy Research
Centre (Speirs et al., 2013) quantifies the number of publications on
materials availability over the previous decades and confirms the
aforementioned relations as illustrated in Fig. 1.
In addition to research work, several concepts of national
resource strategies such as the “EU Raw Materials Initiative” from
2008 (European Commission, 2008) which was integrated into
national research strategies of EU member states (defra, 2012;
Tiess, 2010) have been published in recent years. In 2013 alone,
the “National Strategic and Critical Minerals Production Act” and
the “National Strategic and Critical Minerals Policy Act” have been
presented to the U.S. House of Representatives. These are two bills
that seek to expedite the development of strategic and critical
160
Number of publications on
materials availability
36
Technology specific aspects of
raw material availability
120
80
Raw material
criticality in general
40
0
1947 -1979 1980 -1989 1990 -1999 2000 -2011
Fig. 1. Number of publications on materials availability since 1947 based on Speirs
et al. (2013). Note that the publications for specific technologies were summarized.
S. Glöser et al. / Resources Policy 44 (2015) 35–46
minerals in the U.S. which confirms the high political relevance of
raw material supply security issues at present.
Basic definitions and methodology of criticality determination
In order to enable a precise derivation of the criticality matrix
approach, some basic definitions and equations are required.
Historically, the terms “critical” and “strategic” have been closely
associated and not clearly differentiated (Haglund, 1984; NRC,
2008). In the current discussion about materials availability, strategic materials are almost exclusively associated with national security and military needs, while supply restrictions of critical materials
form a threat to inflict serious damage on a nation's economy
incorporating all industrial sectors (Evans, 2009). The European
Commission defines raw material criticality as follows (European
Commission, 2010): “To qualify as critical, a raw material must face
high risks with regard to access to it, i.e. high supply risks or high
environmental risks, and be of high economic importance. In such a
case, the likelihood that impediments to access occur is relatively
high and impacts for the whole EU economy would be relatively
significant.”
In this paper, we focus on the concept of the criticality matrix
which is a common tool for the assessment and communication of
raw material criticality. In contrast to simple hierarchical risk rankings,
the “criticality matrix approach”—introduced by the U.S. National
Research Council (NRC, 2008)—follows the definitions above as both
the supply risk and the economic importance of a raw material are
quantified (cf. Fig. 2, right side). Several researchers and policy advisors
have adopted this approach and modified it for further criticality
assessments as discussed in the following sections. However, in some
cases it seems to be overlooked that the criticality matrix is an
abstraction of classical quantitative risk assessment within a risk
matrix, which is widely used in numerous disciplines such as climate
risk management (e.g. Smith, 2013), safety engineering (e.g. Salvi and
Debray, 2006), or project risk management (e.g. Smith et al., 2014).
When regarding the simplest case of a binary possibility of “damage”
or “no damage”, risk is defined as the product of probability of
37
occurrence of a specific scenario and the consequences caused by this
scenario (cf. International Standards Organization, 2009). Hence, in
this simplified consideration, risk is a measure for the expected
damage caused by a specific scenario. Eq. (1) summarizes different
quantitative risk definitions (Cox, 2008; Kaplan and Garrick, 1981;
Webster, 2011).
Risk ¼ probability of occurrence consequence
¼ likelihood vulnerability
¼ relative frequency severity
ð1Þ
The visualization and comparison of potential events is often displayed
in a risk matrix—also referred to as a probability impact grid (PIG)—as
illustrated in Fig. 2.
In the case of raw material criticality determination, the supply
risk, which aims to reflect the likelihood of supply disruptions, is
plotted against the vulnerability due to supply disruptions which
can be interpreted as a measure of the economic importance of a
raw material with consideration of potential direct substitution,
hence, the consequences for an economy in case of supply disruptions (cf. Fig. 2). Since neither the economic importance of a raw
material nor its supply risk can be unambiguously measured,
different indicators are used as proxies in order to quantify the
two dimensions of “supply risk and vulnerability”.
Following the classical risk definition in Eq. (1), Eq. (2) defines
raw material criticality as the product of the likelihood of supply
disruptions and their economic consequences:
Raw material criticality ¼ supply risk vulnerability
¼ likelihood of supply disruptions economic consequences
ð2Þ
Taking the definition of raw material criticality as described in Eq.
(2), a criticality function over the axes of supply risk and vulnerability can be defined (cf. Fig. 3a). By projecting the criticality
function into the two dimensional criticality matrix, each point
within the matrix gets a specific criticality level in the form of
contour lines of the criticality function (see contour lines in
Fig. 3b). This enables a clear interpretation of each point within
Fig. 2. General risk matrix (also referred to as probability impact grids or risk maps) applied in numerous scientific fields such as in safety and environmental engineering
and the definition of raw material criticality as an abstraction of classical risk assessment the way it was introduced by the U.S. National Research Council (NRC, 2008). In this
context, raw material criticality may be interpreted as the “systemic risk” for an economy due to disturbances in raw material supply. Note that in few cases of the literature
about general risk assessment, the term “criticality matrix” is used as a synonym for “risk matrix” (Asset Insights, 2014; Valbuena, 2010).
38
S. Glöser et al. / Resources Policy 44 (2015) 35–46
Review of selected studies
criticality level 80
criticality level 60
criticality level 40
criticality level 20
10
Vulnerability (consequence)
9
8
7
6
5
4
3
2
1
0
0
1
2
3
4
5
6
7
8
9
10
Supply risk (likelihood)
Fig. 3. Criticality function with contour lines in the context of classical risk
definition (cf. Glöser and Faulstich, 2014). Raw material criticality in this concept
is defined as follows: criticality ¼vulnerability supply risk. The contour lines
represent different combinations of “supply risk” and “vulnerability” that lead to
the same criticality level. While in the logarithmically scaled risk matrix, levels of
uniform risk (likelihood consequence) are parallel, in the case of the linearly
scaled criticality matrix the levels of uniform criticality exhibit a convex shape. The
scale of the axes between 0 and 10 was chosen arbitrarily and could be replaced by
any other scale. (a) Criticality function derived from the product of the two axes
“vulnerability” and “supply risk”. (b) Criticality matrix with contour lines representing the criticality level of the function above.
the matrix because besides the single coordinates, a measure of
criticality level can be assigned to each raw material. Hence, the
uniform risk levels derived from different combinations of “likelihood and consequence” in the risk matrix (cf. risk matrix in
Fig. 2) can be transferred to uniform criticality levels through
different combinations of “supply risk and vulnerability” (cf.
criticality matrix in Fig. 3b). Note that in classical risk matrices
the scaling of the axes is generally logarithmic, which results in
linear contour lines with less convex character (cf. squares of
uniform risk levels within the risk matrix in Fig. 2). However,
because none of the current criticality studies used logarithmic
scaling for their matrices, the criticality function is defined with a
linear scaling resulting in convex contour lines (cf. Fig. 3b).
Before comparing and discussing the results of major criticality
studies in the context of the criticality definition shown in Fig. 3,
several selected approaches of criticality determination are shortly
described in this section in order to establish an overview of relevant
results and indicator aggregations which is helpful in understanding
the following discussion. Thereby, we focus on studies analyzing
larger numbers of raw materials and defining thresholds between
critical and non-critical materials or overall criticality levels which are
then compared with the contour lines derived from a uniform
criticality matrix as described in Fig. 3b (cf. section “Results and
discussion”). Studies using the criticality matrix as a more qualitative
tool to visualize and communicate supply risks and vulnerabilities
(e.g. Duclos et al., 2008; NRC, 2008; U.S. DoE, 2010, 2011) are of minor
relevance for the discussion in this paper. An overview of these
studies is provided in the supplementary data accompanying this
paper. The criticality matrix approach described above was particularly used for the quantification and communication of raw material
criticality on national levels including the European Union. However,
the criticality matrix was also applied on the company and the global
level. Table 1 provides an overview of major studies based on the
concept of criticality matrices.
As described in Fig. 2, the principle of criticality determination
within a criticality matrix was introduced by the U.S. National
Research Council (NRC, 2008). The quantification of the supply
risk and the economic importance in this study was realized (by
the experts of the committee) through the combination of
quantitative indicators that formed the basis for qualitative
expert judgment (a summary of this method is provided in the
accompanying supplementary data). In the NRC study, 13 different raw materials or raw material groups were analyzed and
compared (cf. Fig. 8, left side).
In 2010, the European Commission presented a quantitative
approach of criticality determination (European Commission, 2010)
using a criticality matrix as proposed by the NRC (2008). In a similar
manner to the NRC study, this study was prepared (with external
support) by an expert committee appointed by the European
Commission, the “Ad Hoc Working Group on Defining Critical Raw
Materials”. The committee analyzed 41 industrial metals and minerals regarding their economic importance for the EU industry and
their supply risk. The quantification of the coordinates of each raw
material within the criticality matrix (economic importance and
supply risk) was calculated by several equations using quantitative
indicators. Thus, this approach is less dependent on expert judgment
but focuses on clearly quantifiable figures, enhancing the transparency of the coordinates within the criticality matrix. Nonetheless, the
choice and aggregation of the indicators necessarily incorporates
personal judgments of the commissioners and remains—as all multiple indicator methods—open to debate.
The EU-method of calculating the supply risk for each raw
material is provided in Eq. (3) (HHI as the Herfindahl–Hirschmann-Index).
X
X 2
Supply riski ¼ ð1 ρi Þ ðAis σ is Þ ðP ic ½WGIc or EPIc Þ
ð3Þ
|fflfflfflffl{zfflfflfflffl}
s
c
Recycability |fflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
Substitutability
WGI or EPI weighted HHI
The economic importance is based on the total value added of the
production sectors (mega sectors) that depend on the specific raw
material and was calculated as described in the following equation:
P
ðS A Þ
Economic importancei ¼ s is is
ð4Þ
GDPEU27
Table 2 provides a definition and explanation to the underlying
indicators in Eqs. (3) and (4).
S. Glöser et al. / Resources Policy 44 (2015) 35–46
39
Table 1
Major studies which were carried out within the past 7 years in order to quantify raw material criticality using the concept of a criticality matrix as introduced by the NRC
(cf. Achzet and Helbig, 2013; Erdmann and Graedel, 2011; Gandenberger et al., 2012; Speirs et al., 2013).
Title
Focus
Commissioners/reference
Criticality matrix methodology as introduced by the NRC (cf. Fig. 2)
“Minerals, Critical Minerals, and the U.S. Economy”
U.S. economy
“Design in an Era of Constrained Resources”
GE corporation
“Critical raw materials for the EU”
European Union
“Critical Materials Strategy”
Clean energy technologies
“Critical Raw Materials for Germany”
German industry
National Research Council (NRC, 2008)
General Electric (Duclos et al., 2008)
European Commission (European Commission, 2010, 2014)
U.S. Department of Energy (U.S. DoE, 2010, 2011)
KfW bank group (Erdmann et al., 2011)
Criticality space (matrix enhanced by the third dimension of “environmental implications”)
“Criticality of the Geological Copper Family”
U.S. economy and global level
“Criticality of Iron and its Principal Alloying Elements”
U.S. economy and global level
“Criticality of the Geological Zinc, Tin and Lead Family”
U.S. economy and global level
Yale University (Nassar et al., 2012)
Yale University (Nuss et al., 2014)
Yale University (Harper et al., 2015)
Table 2
Indicators used by the “Ad hoc Working Group on defining critical raw materials” in order to quantify the supply risk (cf. Eq. (3)) and the economic importance of a raw
material (cf. Eq. (4)). (European Commission, 2010).
“Supply risk” dimension
“Economic importance” dimension
ρi – share of postconsumer recycling material in total material use in production (EoL Recycling Input Rate)
Sis – total value added of sector s depending on
raw material i.
Ais – share of raw material i used in sector s
(identical to supply risk)
Ais – share of raw material i used in sector s
σis – substitutability of raw material i in sector s
Pic – share of country c in the global production of raw material i
GDPEU27 – overall GDP of the EU27
WGIc – value of the World Governance Indicators (Kaufmann et al., 2013) for country c (scaled such that a higher score
corresponds to poorer governance)
EPIc – value of the Environmental Performance Index (Yale Center for Environmental Law and Policy, 2014) for country
c (scaled such that a higher score corresponds to poorer environmental performance)
Fig. 4. Results of the current EU criticality study in the form of the criticality matrix published by the “Ad Hoc Working Group on Defining Critical Raw Materials”, an expert
committee appointed by the European Commission (European Commission, 2014). 20 critical raw materials for the European economy have been identified. However,
regarding the concept of criticality in the context of a risk matrix, the course of the threshold between critical and non-critical materials is questionable (cf. Fig. 9b in the next
section).
Unlike the NRC (2008) study which aimed to compare the
criticality of raw materials (“less critical” vs. “more critical”), the
EU study aimed to generate a list of critical raw materials for the
EU as an instrument for both communication and policy making.
Thus, a decision was made by the Ad-hoc Working Group on the
introduction of fixed thresholds for both the economic importance
and supply risk dimensions, defining a rectangular “critical region”
within the criticality matrix (we explore below how the shape of
this region is at odds with classical risk assessment).
An update of the EU criticality matrix was published in May 2014
(European Commission, 2014). There were no changes in the basic
methodology used for the updated criticality assessment compared
to the EU 2010 study: both the equations used and the thresholds
set to distinguish between “critical” and “non-critical” raw materials
remained the same. However, the scope was extended to cover 54
raw materials (splitting the rare earth elements into “light” and
“heavy”, including some biotic materials as well as some additional
metals and minerals). The current criticality matrix for the EU and
the threshold that separates the 20 critical raw materials from the
rest of the analyzed commodities is shown in Fig. 4.
The additional project report of the EU 2014 study (Chapman
et al., 2013) contains a further version of the criticality matrix in
40
S. Glöser et al. / Resources Policy 44 (2015) 35–46
which the supply risk is either calculated with the World Governance Indicator (WGIc) or the Environmental Performance Index (EPIc)
of each production country in order to take into account the
environmental risk of primary production (cf. Eq. (3) and Table 2).
The maximum of the respective values derived from Eq. (3) was
finally used as the supply risk measure. This approach gets closer to
the definition of critical raw materials by the European Commission
which incorporates environmental risks in the supply risk discussion
(cf. basic definitions in Section “Basic definitions and methodology
of criticality determination”). Furthermore, Chapman et al. (2013)
present several suggestions for methodological improvements to the
EU approach such as the incorporation of substitutability into the
“Economic Importance” measure (which would come closer to the
concept of vulnerability) and the introduction of curved thresholds
(which would come closer to classical risk assessment, see section
“Results and discussion”) and a transition zone between “critical”
and “non-critical” raw materials.
A study about critical raw materials for the German industry was
carried out by Erdmann et al. (2011) on behalf of the KfW Banking
Group (German government-owned development bank). The methodology applied by the authors is also based on the criticality matrix
as introduced by the NRC (2008). The calculation of the coordinates
incorporates multiple indicators which were differently weighted.
The authors assigned values between 0 and 1 to all indicators,
mainly based on quantitative data but also incorporating qualitative
aspects such as the sensitivity of affected value chains within the
German industry. Table 3 displays the choice and weighting of the
indicators for the KfW study.
As opposed to the EU studies which only distinguish between
critical and non-critical materials through a fixed threshold
(European Commission, 2010, 2014), Erdmann et al. (2011) divided
the criticality matrix into 6 different sections as shown in Fig. 5.
The different criticality sections are defined as follows:
I.
II.
III.
IV.
V.
VI.
Low criticality level (low supply risk, low vulnerability).
Low supply risk, high vulnerability.
High supply risk, low vulnerability.
Medium criticality (medium supply risk, medium vulnerability).
High criticality (high supply risk, high vulnerability).
Maximum criticality (maximum supply risk, maximum vulnerability).
Furthermore, the study includes a sensitivity analysis regarding
the effect of indicator weighting on the overall results. The indicators
are divided into short-term indicators and mid- to long-term indicators (see Table 3). In addition to the general criticality matrix shown
in Fig. 5 (based on the indicator weighting in Table 3), this enabled a
further analysis of the criticality matrix with different time horizons
by incorporating either only short-term indicators or only mid- to
long-term indicators into the criticality calculation. However, one has
to bear in mind that even though all results are presented within the
same criticality matrix, the underlying indicators are different, hence
there is only limited comparability. In order to get a real understanding of the development of raw material criticality over time, all
underlying indicators should be analyzed over time, which in most
cases is connected with a high degree of uncertainty.
Graedel et al. (2012) developed a methodology of criticality determination within a “criticality space” which is based on the axes of
the criticality matrix described by the NRC and enhanced by a third
dimension of “environmental implications” (Nassar et al., 2012). This
indicator quantifies impacts to human health and to the ecosystem
evaluating life cycle inventory data from different databases such as
the “Ecoinvent” project (Swiss Center of Life Cycle Inventory, 2014).
Furthermore, the dimensions of “vulnerability to supply restriction”
and “supply risk”—besides the World Governance Indicators (WGI)
published by the World Bank—incorporate several further subindicators from different sources such as the Policy Potential Index
(PPI) (Wilson et al., 2013), the Human Development Index (HDI)
(Malik, 2013) and the Global Innovation Index (GII) (Dutta, 2013) in
order to quantify the performance of raw material producing and
consuming countries. The authors distinguish between raw material
criticality at a corporation, a national and a global level. The assessment of the coordinates of the criticality space on a national level for
the U.S. including the weighting of the different sub-indicators is
summarized in Fig. 6.
The overall criticality level within the criticality space is
defined as the length of the vectors between the origin and the
metal's location in the criticality space:
overall criticalityi ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
2
2
vulnerabilityi þ supply riski þenvironmental implicationsi
ð5Þ
This methodology of criticality determination has been applied to
the geological copper family (Graedel et al., 2012), iron and its
principal alloying elements (Nuss et al., 2014) and the geological
zinc, tin and lead family (Harper et al., 2015). The results of the
study about the geological copper family including the overall
criticality level for the U.S. in the form of the corresponding
vector lengths are illustrated in Fig. 7. The study further included
a stochastic uncertainty evaluation with respect to data uncertainties which results in a cloud of potential positioning around
the mean values of the coordinates.
Results and discussion
Recent publications in the field of raw material criticality
determination analyzed and discussed the indicators used for the
quantification of supply risks and vulnerability to supply restrictions (Achzet and Helbig, 2013; Erdmann and Graedel, 2011).
Herein, we focus on the interpretation of the results of criticality
assessments within the criticality matrix in the context of classical
risk analysis. Therefore, we used the criticality function described in
Table 3
Indicators and indicator weighting in the KfW study analyzing raw material criticality for the German industry (Erdmann et al., 2011).
Vulnerability
Relevance of volume
German share of global use (2008)
Change of German share of global use (2004–2008)
Change of import volume (2004–2008)
Strategic relevance
Sensitivity of affected value chains in Germany
Push in global demand due to emerging technologies (2030)
Substitutability
P
Time horizon
%
Short term
Short term
Short term
25
10
10
Mid to long term
Mid to long term
Mid to long term
25
20
10
100
Supply risk
Country risk
Country risk of German imports
Country risk of global production
Country concentration of global reserves
Market risk
Concentration of production companies
Static depletion time (2008)
Structural risk
Byproduct ratio
Recycling potential
P
Time horizon
%
Short term
Short term
Short term
10
10
10
Short term
Mid to long term
25
25
Mid to long term
Mid to long term
100
10
10
S. Glöser et al. / Resources Policy 44 (2015) 35–46
41
Fig. 5. Results of the criticality study commissioned by the KfW Banking Group for the German industry (Erdmann et al., 2011). In contrast to the EU studies, different levels
of criticality are distinguished by dividing the matrix into several sections.
Fig. 6. Indicator aggregation for the calculation of the three axes of criticality
space: supply risk (medium-term national level), vulnerability to supply restrictions (national level), and environmental implications (Nassar et al., 2012).
Fig. 3 to compare different criticality studies based on the criticality
matrix concept.
Comparison of major approaches
A direct comparison of the different positions within the matrix
across different studies becomes possible by projecting the results
of the aforementioned studies into a uniform criticality matrix
with contour lines. Via the criticality level (contour lines) a clear
hierarchical list of critical materials comparable to the results of
single index studies can be further extracted from the matrix (see
bar graph with different colors representing the criticality level in
Fig. 8, right side). In order to transfer the original matrices into the
uniform criticality matrix described in Fig. 3, the axes and the
materials’ coordinates were normalized to values from 0 to 10. The
transformation of the original criticality matrix into the uniform
matrix is illustrated in Fig. 8 using the NRC (2008) study as an
example.
Fig. 7. Results of the study “Criticality of the Geological Copper Family” carried out
by Graedel et al. (2012) at a national level (U.S. economy). The criticality is no
longer assessed within the criticality matrix but within a criticality space taking
into account the third dimension of environmental implications. The overall
criticality of a raw material is defined through the vector length as illustrated by
the dashed lines.
Differences in choice and weighting of indicators in order to
quantify the economic importance of a raw material or the supply
risk will always remain subject to the individual judgment of the
persons who carry out the criticality study (Lloyd et al., 2012). Hence,
besides the differences in the systems being analyzed (regions, industries, time frames, etc.), individual risk awareness of the authors
will be included in any criticality study. However, once the material's
position within the criticality matrix is found, the interpretation of
the results should be clear.
Fig. 9 shows the transformation described above (cf. Fig. 8) to the
results of the EU criticality studies (European Commission, 2010,
2014). Inspection of Fig. 9 reveals that the thresholds which were
chosen to separate critical raw materials from non-critical raw
materials are not in line with the contour levels of the criticality
function. This can lead to significant misinterpretation. For example,
in the current EU criticality matrix both chromium and borate are
directly located on the threshold to the “criticality zone” (cf. Fig. 9b),
42
S. Glöser et al. / Resources Policy 44 (2015) 35–46
Fig. 8. Principle and results of transferring the NRC criticality matrix (NRC, 2008) into a uniform matrix with contour lines. The original NRC criticality matrix (left side) is
normalized to axis values from 0 to 10 and then transferred into the uniform criticality matrix described in Fig. 3. (For interpretation of the references to color in this figure
caption, the reader is referred to the web version of this paper.)
which gives the impression that both materials just reach the criteria
to be considered as critical. In fact, both materials have the same
score of supply risk but chromium has a much higher score regarding
its economic importance, thus, a higher criticality level.
When comparing the criticality levels (contour lines) in Fig. 9b,
vanadium is not considered critical even though its criticality level
is above the level of phosphate rock and borate which are both
declared as critical materials by the European Commission. The
same is true when regarding the contour lines in the EU 2010 study
(cf. Fig. 9a). While cobalt was considered critical, chromium was
not, even though the criticality level of chromium in the EU 2010
study is above the level of cobalt. In addition to the criticality
matrix, the bar plots in Fig. 9 illustrate a hierarchy of the most
critical raw materials in the EU studies regarding their criticality
level. While the U.S. National Research Council introduced the
criticality matrix as a more qualitative tool to communicate and
visualize economic vulnerabilities due to insecure raw material
supply, the EU studies implement a quantitative approach to
analyzing and comparing a large array of different raw materials.
Especially in these studies it is imperative to insure a quantitatively
correct data interpretation. In this context, it seems reasonable to
communicate the 10, 15 or 20 most critical metals and minerals for
the European industry (which could be realized through the
criticality levels) rather than to simply distinguish between critical
and non-critical materials through fixed thresholds which are not in
line with basic risk definitions.
The different criticality categories defined in the KfW study for
Germany (Erdmann et al., 2011) are, at least for the top two
“criticality groups”, closer to the risk based definition of raw material
criticality. However, when comparing the thresholds in the original
study (cf. Fig. 5) with the contour lines from the criticality function
(cf. Fig. 3 ), several mismatches can be identified as illustrated in
Fig. 10. For example, tungsten and antimony are almost on the same
criticality level, but tungsten is assigned to the “high criticality group”
while antimony is accounted to the “maximum criticality group”
(cf. Fig. 10). Similarly, tellurium is located on a higher contour line
than chromium and niobium but tellurium is assigned to the
“medium criticality group” while chromium and niobium are considered as “highly critical”.
Convex vs. concave criticality levels
Figs. 9 and 10 illustrate how thresholds used in major studies
and defined as rectangles (European Commission, 2010, 2014) or
polygons (Erdmann et al., 2011) are not in line with classical risk
assessment. The incongruence becomes striking when defining raw
material criticality as the vector length (Euclidean distance)
between the origin and the material's position in the criticality
matrix or the criticality space (cf. Fig. 11). In this case, the overall
criticality level is derived by summation of different indicators
(cf. Eq. (5)) and not through a product in the context of consequence
and likelihood (cf. Eq. (2)). Following this approach, a raw material
might be considered critical due to a high likelihood of supply
disruptions (high supply risk) even though the vulnerability level is
very low (cf. metal B on the right side of Fig. 11). Fig. 11 clarifies the
differences in resulting contour lines (positions of uniform criticality levels) when overall criticality is defined as an abstraction of
the classical risk definition resulting in convex contour lines (matrix
in the middle) or through the vector length resulting in concave
contour lines (matrix on the right side).
Furthermore, when including the additional dimension of
“Environmental Implications” into the analysis of criticality (cf.
Graedel et al., 2012), it has to be considered that the environmental risk during mining and refining of raw materials and the
supply risk are not independent from each other. That is, a metal
or a group of metals will face higher supply risks if the mining and
primary processing cause severe environmental impacts, not least
because environmental and human health issues are gaining
higher attention in developing and emerging countries which
are responsible for the majority of global mining activities
(Behrens et al., 2007). In this context, the suggestion in the EU
2014 study (cf. Section “Review of selected studies”) to incorporate
the environmental performance in the calculation of supply risk
appears reasonable. Hence, the criticality matrix in the context of a
classical risk matrix and the criticality space using the vector
length between the origin and a metal's position in space (cf.
Fig. 11, left side) as the criticality measure are very different
approaches and should not be confused. Nonetheless, vector
length calculations may be used for the visualization of the
aggregation of specific indicators such as a sustainability index of
raw material production, taking into account the 3 dimensions of
sustainability (economic, social and environmental aspects). However, these indicator aggregations are different from the original
concept of raw material criticality in the context of risk assessment
in the way it has been introduced by the U.S. National Research
Council (NRC, 2008). The general inconsistency of Euclidean
distances (vector lengths) for comparing and assessing risk within
risk matrices was pointed out by Webster (2011).
S. Glöser et al. / Resources Policy 44 (2015) 35–46
43
Fig. 9. Results of the two EU matrices within the uniform scaling. Note that the axes were changed compared to the original matrix in Fig. 4 (vertical: economic importance,
horizontal: supply risk). The original threshold (dashed red lines cf. Fig. 4) traverses numerous criticality levels thus declaring materials with the same criticality level
sometimes critical, sometimes non-critical. (a) Projection of the EU 2010 results into the uniform criticality matrix with contour lines. (b) Projection of the EU 2014 results
into the uniform criticality matrix with contour lines. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this
paper.)
Conclusions and recommendations for further research
The concept of criticality determination within a criticality matrix
is a powerful tool to identify and communicate economic vulnerabilities due to insecure raw material supply. This approach appears
particularly useful as a screening method to analyze large numbers of
different materials and to compare the relative supply situation
among different industrial raw materials in order to identify those
materials for which supply disruptions pose the largest threat to an
economy. Therefore, it is reasonable to incorporate a whole range of
industrial metals and minerals in the criticality assessment the way it
was performed by the European Union (European Commission, 2010,
2014) or by the KfW bank group (Erdmann et al., 2011). The analysis
of the criticality of a small group of metals does not seem very
helpful for decision makers as the absolute value of criticality
assigned to an individual material determined with specific indicator
weightings has only limited informative value.
When comparing the aforementioned criticality matrix determinations, two major ways in which they differ can be identified:
the choice and weighting of different indicators in order to
quantify the two axes of the criticality matrix (supply risk and
vulnerability),
the interpretation of a material's coordinates within the criticality matrix.
While the first item (choice & weighting of indicators) will
always be influenced by the individual judgment of the persons
being involved in the criticality determination, the interpretation of
the position of a raw material within the criticality matrix should be
(more) uniform across studies, especially considering their common
intention and roots in classical risk assessment. However, as shown
in this paper, the close relation between the criticality matrix and
classical risk analysis within a risk matrix has not been sufficiently
44
S. Glöser et al. / Resources Policy 44 (2015) 35–46
Fig. 10. Results of the KfW study for Germany within the uniform criticality matrix. The dashed red lines show the original thresholds of the different criticality sections (cf.
Fig. 5). (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)
Fig. 11. Raw material criticality measure through the product of supply risk and vulnerability and vector lengths (Euclidean distances). Note that in the case of vector length
measure a raw material may be considered critical even though only one of the axis values is high. Generally, distance measures in risk matrices are not in line with the
original definition of risk (see Webster, 2011).
taken into account in most studies. Vigorous efforts in technical
research regarding material substitution, material efficiency and
recycling are taken based on the results of these criticality studies.
This is particularly true when regarding the political impact of the
EU criticality studies (European Commission, 2010, 2014) which are
considered in numerous national research strategies. Therefore,
besides a justified indicator selection and weighting (cf. Achzet
and Helbig, 2013), a consistent interpretation of a material's position within the criticality matrix is indispensable. Taking into
account the close relation between the criticality matrix and the
classical risk matrix is a prerequisite for the correct interpretation
and communication of future quantitative criticality assessments
based on the matrix concept.
However, there are several further aspects which should be
considered in future research work. Regarding the effect of uncertainties, analyses should not be reduced not only to data uncertainties but also to variations in indicator weightings which can be
determined with sensitivity analyses as suggested by Erdmann and
Graedel (2011). Apart from the discussion about choice and weighting of suitable indicators to quantify raw material criticality, the
dimension of time should be given stronger attention in future
studies. The previous studies on raw material criticality are static
analyses which assess today's supply situation and do not include
any dynamic aspects. Some studies take indicators into account that
provide information about mid- and long-term supply risks (cf.
Table 3). However, by projecting indicators from different points in
time (mid-term and long-term future) to the present without
considering the dimension of time, the results might be diluted.
That is, one raw material might be considered critical due to current
market situations (referring to short-term indicators which are
certain at present) while an other material might reach a high
criticality level due to expected future developments (based on midto long-term indicators which might change over time and are linked
to uncertainties).
For decision makers from policy and industry, detailed analyses of
material flows within the system in focus have to be performed to
S. Glöser et al. / Resources Policy 44 (2015) 35–46
establish a sufficient database for the comparison of potential measures in order to improve the supply situation. In this context, dynamic
material and substance flow models that quantify material flows
within clearly specified spatial boundaries are a prerequisite to obtain
a better understanding about recycling potentials, resource efficiency,
import dependencies and sectoral use patterns along the entire value
chain (e.g. Buchner et al., 2014; Glöser et al., 2013; Peiró et al., 2013). A
comprehensive review of major approaches of dynamic material flow
models has recently been presented by Müller et al. (2014) and
Bornhöft et al. (2013) analyzed suitable software environments for
dynamic material flow modeling. First dynamic approaches of criticality determination using different methods such as agent based
modeling (Knoeri et al., 2013), system dynamics modeling (Glöser and
Faulstich, 2012) or scenario based risk analysis (Roelich et al., 2014)
have lately been introduced. An advantage of these model based
approaches is the possibility to simulate numerous scenarios and
perform cumulative risk analysis including stochastic models, which is
in contrast to the binary combination of supply risk and vulnerability
in the static criticality studies. However, these approaches are no
longer screening methods analyzing a large number of materials, but
very specific models for individual substances which have been
identified as critical in previous screening analyses.
Acknowledgments
The work leading to this paper was supported by the AERTOs
Community (Associated European Research and Technology Organizations) and by the Office of Technology Assessment at the
German Bundestag (TAB) which is gratefully acknowledged. The
helpful feedback of an anonymous reviewer is also kindly noted.
The authors would further like to particularly thank Matthias Pfaff,
Katharina Eckartz, Anna Stange and Stefan Haag for their support
in securing relevant data and correcting the paper. However,
statements presented here are not the official opinion/position of
these colleagues. The responsibility for the content of this paper
remains with the authors.
Appendix A. Supplementary data
In addition to this paper, in the supplementary data we provide
an enhanced overview of relevant criticality studies using the
principle of a criticality matrix.
Supplementary data associated with this paper can be found in
the online version at http://dx.doi.org/10.1016/j.resourpol.2014.12.003.
References
Achzet, B., Helbig, C., 2013. How to evaluate raw material supply risks—an
overview. Resour. Policy 38 (4), 435–447.
Alonso, E., Gregory, J., Field, F., Kirchain, R., 2007. Material availability and the
supply chain: risks, effects, and responses. Environ. Sci. Technol. 41 (19),
6649–6656.
APS, 2011. Energy Critical Elements: Securing Materials for Emerging Technologies:
A Report by the APS Panel on Public Affairs & the Materials Research Society.
Technical Report, MRS.
Asset Insights, 2014. Critcality Matrix (Risk Matrix). Online. URL ⟨http://www.
assetinsights.net/Glossary/G_Criticality_Matrix.html⟩.
Behrens, A., Giljum, S., Kovanda, J., Niza, S., 2007. The material basis of the global
economy: worldwide patterns of natural resource extraction and their implications for sustainable resource use policies. Ecol. Econ. 64 (2), 444–453, Special
Section—Ecosystem Services and Agriculture Ecosystem Services and
Agriculture.
Bleischwitz, R., Dittrich, M., Pierdicca, C., 2012. Coltan from Central Africa, international trade and implications for any certification. Resour. Policy 37 (1), 19–29.
Bornhöft, N.A., Nowack, B., Hilty, L.M., 2013. Material flow modelling for environmental exposure assessment—a critical review of four approaches using the
comparative implementation of an idealized example. In: Proceedings of the
27th EnviroInfo Conference 2013, Hamburg, Germany, pp. 379–388.
45
Buchner, H., Laner, D., Rechberger, H., Fellner, J., 2014. In-depth analysis of
aluminum flows in austria as a basis to increase resource efficiency. Resour.
Conserv. Recycl. 93 (0), 112–123.
Buijs, B., Sievers, H., Tercero Espinoza, L.A., 2012. Limits to the critical raw materials
approach. Proc. ICE-Waste Resour. Manag. 165 (4), 201–208.
Campbell, G.A., 2014. Rare earth metals: a strategic concern. Miner. Econ. 27 (1),
21–31.
Chapman, A., Arendorf, J., Castella, T., Tercero Espinoza, L., Klug Stefan, Wichmann,
E., 2013. Study on Critical Raw Materials at Eu Level: Final Report. Technical
Report, Oakdene Hollins, Fraunhofer ISI.
Commission of the European Communities, 1975. The Community's Supplies of Raw
Materials. Technical Report, Commission of the European Communities, Brussels, Belgium. URL ⟨http://aei.pitt.edu/1481/⟩.
Committee on Science, 1984. The National Critical Materials Act of 1984. Technical
Report on Public Law 98:373, Committee on Science and Technology, House
of Representatives, Washington, DC. URL ⟨https://archive.org/details/
nationalcritical00unit⟩.
Congressional Budget Office, 1983. Strategic and Critical Nonfuel Minerals: Problems and Policy Alternatives. Technical Report, Congressional Budget Office,
Congress of the United States, Washington, DC.
Cox, A.L., 2008. What's wrong with risk matrices? Risk Anal. 28 (2), 497–512.
defra, 2012. A Review of National Resource Strategies and Research. Technical
Report, Department for Environment Food and Rural Affairs, London, UK.
Duclos, S.J., Otto, J.P., Konitzer, D.G., 2008. Design in an era of constrained resources.
Mech. Eng. 132, 36–40.
Dutta, Soumitra; Lanvin, B., 2013. The Global Innovation Index 2013: The Local
Dynamics of Innovation. Johnson Cornell University, Ithaca, NY.
Erdmann, L., Behrendt, S., Feil, M., 2011. Kritische Rohstoffe für Deuschland,
Anhang: Identifikation aus Sicht deutscher Unternehmen wirtschaftlich
bedeutsamer mineralischer Rohstoffe, deren Versorgungslage sich mittel- bis
langfristig als kritisch erweisen könnte. Technical Report, IZT/adelphi, Berlin.
Erdmann, L., Graedel, T.E., 2011. Criticality of non-fuel minerals: a review of major
approaches and analyses. Environ. Sci. Technol. 45 (18), 7620–7630.
European Commission, 2008. The Raw Materials Initiative—Meeting Our Critical
Needs for Growth and Jobs in Europe. Technical Report, European Commission,
Brussels, Belgium.
European Commission, 2010. Critical Raw Materials for the EU. Technical Report,
European Commission (Enterprise and Industry), Brussels, Belgium.
European Commission, 2014. Report on Critical raw materials for the EU. Technical
Report, European Commission (Enterprise and Industry), Brussels, Belgium.
Evans, A.M., 2009. Ore Geology and Industrial Minerals: An Introduction, 3rd
edition Blackwell Scientific Publications, Oxford.
Fizaine, F., 2013. Byproduct production of minor metals: threat or opportunity for
the development of clean technologies? the pv sector as an illustration. Resour.
Policy 38 (3), 373–383.
Gandenberger, C., Glöser, S., Marscheider-Weidemann, F., Ostertag, K., Walz, R.,
2012. Die Versorgung der deutschen Wirtschaft mit Roh- und Werkstoffen für
Hochtechnologien: Präzisierung und Weiterentwicklung der deutschen Rohstoffstrategie: Innovationsreprot. Arbeitsbericht, vol. 150. Büro für Technikfolgenabschätzung beim deutschen Bundestag, Berlin.
Glöser, S., Faulstich, M., 2014. Analyse kritischer Rohstoffe durch Methoden der
Multivariaten Statistik. In: Teipel, U., Reller, A. (Eds.), 3. Symposium Rohstoffeffizienz und Rohstoffinnovationen. Fraunhofer Verlag, Stuttgart, Germany,
pp. 53–79.
Glöser, S., Faulstich, M., 2012. Quantitative analysis of the criticality of mineral and
metallic raw materials based on a system dynamics approach. In: Proceedings
of the 30th International Conference of the System Dynamics Society, St. Gallen,
Switzerland 2012. System Dynamics Society, Albany, NY.
Glöser, S., Soulier, M., Tercero Espinoza, L.A., Faulstich, M., 2013. Using dynamic
stock and flow models for global and regional material and substance flow
analysis. In: Proceedings of the 31st International Conference of the System
Dynamics Society, Cambridge, MA. System Dynamics Society, Albany, NY.
Graedel, T.E., Barr, R., Chandler, C., Chase, T., Choi, J., Christoffersen, L., Friedlander, E.,
Henly, C., Jun, C., Nassar, N.T., Schechner, D., Warren, S., Yang, M.-Y., Zhu, C., 2012.
Criticality of the geological copper family: methodology of metal criticality
determination. Environ. Sci. Technol. 46 (2), 1063–1070.
Graedel, T.E., Nassar, N.T., 2013. The Criticality of Metals: A Perspective for
Geologists. Geological Society, London, Special Publications.
Haglund, D.G., 1984. Strategic minerals. Resour. Policy 10 (3), 146–152.
Harper, E., Kavlal, G., Burmeister, M., Erbis, S., Espinoza, V., Nuss, P., Graedel, T., 2015.
Criticality of the geological zinc, tin, and lead family. J. Ind. Ecol, http://dx.doi.
org/10.1111/jiec.12213.
Hoenderdaal, S., Tercero Espinoza, L., Marscheider-Weidemann, F., Graus, W., 2013.
Can a dysprosium shortage threaten green energy technologies? Energy 49,
344–355.
Humphreys, D., 1995. Whatever happened to security of supply? minerals policy in
the post-cold war world. Resour. Policy 21 (2), 91–97.
Humphreys, D., 2010a. The great metals boom: a retrospective. Resour. Policy 35
(1), 1–13.
Humphreys, D., 2010b Minerals: Industry History and Fault Lines of Conflict.
Polinares Working Paper no. 4, 2010, pp. 1–27.
International Standards Organization, 2009. Risk Management—Principles and
Guidelines. URL ⟨http://www.iso.org/iso/catalogue_detail?csnumber=43170⟩.
Jacobson, D.M., Turner, R.K., Challis, A., 1988. A reassessment of the strategic
materials question. Resour. Policy 14 (2), 74–84.
46
S. Glöser et al. / Resources Policy 44 (2015) 35–46
Kaplan, S., Garrick, J.B., 1981. On the quantitative definition of risk. Risk Anal. 1.1 (1),
11–27.
Kaufmann, D., Kraay, A., Mastruzzi, M., 2013. The Worldwide Governance Indicators
(WGI). Technical Report, World Bank, ⟨http://info.worldbank.org/governance/
wgi/index.aspx#home⟩.
Kesicki, F., 2010. The third oil price surge-what's different this time? Energy Policy
38 (3), 1596–1606.
Knoeri, C., Wäger, P.A., Stamp, A., Althaus, H.-J., Weil, M., 2013. Towards a dynamic
assessment of raw materials criticality: linking agent-based demand – with
material flow supply modelling approaches. Sci. Total Environ. 461–462,
808–812.
Le Billon, P., 2001. The political ecology of war: natural resources and armed
conflicts. Polit. Geogr. 20 (5), 561–584.
Leamy, H., 1985. National critical materials act: call for involvement in new council
designated as national watchdog of materials research activities. MRS Bull.
1985, 21–25.
Legislative Councel, 1939. Strategic and Critical Materials Stock Piling Act. Technical
Report, 50 U.S.C. 98, Office of the Legislative Council, U.S. House of Representatives, Washington, D.C. URL ⟨http://legcounsel.house.gov/Comps/Strategic%
20And%20Critical%20Materials%20Stock%20Piling%20Act.pdf⟩.
Lloyd, S., Lee, J., Clifton, A., Elghali, L., France, C., 2012. Recommendations for
assessing materials criticality. Proc. ICE-Waste Resour. Manag. 165 (4), 191–200.
Malik, K., 2013. Human Development Report: The Rise of the South: Human
Progress in a Diverse World, vol. 2013. United Nations Development Programme, New York, NY.
Mason, E.S., 1952. An american view of raw materials problems: the report of the
president's materials policy commission. J. Ind. Econ. 1 (1), 1–20, URL ⟨http://
www.jstor.org/stable/2097676⟩.
Massari, S., Ruberti, M., 2013. Rare earth elements as critical raw materials: focus on
international markets and future strategies. Resour. Policy 38 (1), 36–43.
Moss, R.L., Tzimas, E., Kara, H., Willis, P., Kooroshy, J., 2011. Critical Metals in
Strategic Energy Technologies: Assessing Rare Metals as Supply Chain Bottlenecks in Low-Carbon Energy Technologies. Technical Report, JRC Scientific and
Technical Reports, European Commission.
Moss, R.L., Tzimas, Willis, P., Arendorf, J., Tercero Espinoza, L., 2013. Critical Metals
in the Path towards the Decarbonisation of the EU Energy Sector: Assessing
Rare Metals as Supply Chain Bottlenecks in Low-Carbon Energy Technologies.
Technical Report, JRC Scientific and Technical Reports, European Commission.
Müller, E., Hilty, L.M., Widmer, R., Schluep, M., Faulstich, M., 2014. Modeling Metal
Stocks and Flows: A Review of Dynamic Material Flow Analysis Methods.
Environmental Science & Technology, 140204134136005.
Nassar, N.T., Barr, R., Browning, M., Diao, Z., Friedlander, E., Harper, E.M., Henly, C.,
Kavlak, G., Kwatra, S., Jun, C., Warren, S., Yang, M.-y., Graedel, T.E., 2012.
Methodology of metal criticality determination: criticality of the geological
copper family. Environ. Sci. Technol. 46 (2), 1071–1078.
Norgate, T.E., Jahanshahi, S., Rankin, W.J., 2007. Assessing the environmental impact
of metal production processes. J. Clean. Prod. 15 (8–9), 838–848.
Novinsky, P., Glöser, S., Kühn, A., Walz, R., 2014. Modeling the feedback of battery
raw material shortages on the technological development of lithium-ionbatteries and the diffusion of alternative automotive drives. In: Proceedings
of the 32nd International Conference of the System Dynamics Society, Delft.
The Netherlands. System Dynamics Society, Albany, NY.
NRC, 2008. Minerals, Critical Minerals, and the U.S. Economy: Committee on Critical
Mineral Impacts of the U.S. Economy, Committee on Earth Resources, National
Research Council. The National Academies Press.
Nuss, P., Harper, E.M., Nassar, N.T., Reck, B.K., Graedel, T.E., 2014. Criticality of iron
and its principal alloying elements. Environ. Sci. Technol. 48 (7), 4171–4177.
Parthemore, C., 2011. Elements of Security: Mitigating the Risks of U.S. Dependence
on Critical Minerals. Center for a New American Security, Washington DC.
Peiró, L.T., Méndez, G.V., Ayres, R.U., 2013. Material flow analysis of scarce metals:
sources, functions, end-uses and aspects for future supply. Environ. Sci.
Technol. 47 (6), 2939–2947.
Robinson, A.L., 1986. Congress critical of foot-dragging on critical materials. Science
234 (1), 20–21.
Roelich, K., Dawson, D.A., Purnell, P., Knoeri, C., Revell, R., Busch, J., Steinberger, J.K.,
2014. Assessing the dynamic material criticality of infrastructure transitions: a
case of low carbon electricity. Appl. Energy 123, 378–386.
Rosenau-Tornow, D., Buchholz, P., Riemann, A., Wagner, M., 2009. Assessing the
long-term supply risks for mineral raw materials—a combined evaluation of
past and future trends. Resour. Policy 34 (4), 161–175.
Salvi, O., Debray, B., 2006. A global view on ARAMIS, a risk assessment methodology for industries in the framework of the SEVESO II directive. J. Hazard. Mater.
130 (3), 187–199.
Sievers, H., Tercero, L., 2012. European Dependence on and Concentration Tendencies of the Material Production. Polinares Working Paper 3/2012 (14).
Smith, K., 2013. Environmental Hazards: Assessing Risk and Reducing Disaster, 6th
edition. Routledge, Abingdon.
Smith, N.J., Jobling, P., Thompson, S., Merna, T., 2014. Managing Risk in Construction
Projects, 3rd edition. Wiley-Blackwell, Chichester, England; Oxford, England.
Speirs, J., Houri, Y., Gross, R., 2013. Materials Availability: Comparison of Material
Criticality Studies: Methodologies and Results. Working Paper III. UK Energy
Research Centre.
Swiss Center of Life Cycle Inventory, 2014. ecoinvent: Lci Database. URL ⟨http://
www.ecoinvent.org/⟩.
Tercero, L., 2012. The Role of Emerging Technologies in Rapidly Changing Demand
for Mineral Raw Materials. Polinares Working Paper.
Tiess, G., 2010. Minerals policy in europe: some recent developments. Resour.
Policy 35 (3), 190–198.
Tilton, J.E., Humphreys, D., Radetzki, M., 2011. Investor demand and spot commodity prices. Resour. Policy 36 (3), 187–195.
U.S. DoE, 2010. Critical Materials Strategy: 2010. Technical Report, U.S. Department
of Energy, Washington DC.
U. S. DoE, 2011. Critical Materials Strategy: 2011. Technical Report, U.S. Department
of Energy, Washington DC.
U.S. Council on International Economic Policy, 1974. Special Report: Critical
Imported Materials. Technical Report, U.S. Council on International Economic
Policy, Executive Office of the President, Washington DC.
Valbuena, G., 2010. Challenges of Deepwater Development. Online. URL ⟨http://
www.offshore-mag.com/articles/print/volume-70/issue-7/subsea/challenge
s-of-deepwater-development.html⟩.
Verhoef, E., Dijkema, G.P., Reuter, M.A., 2004. Process knowledge, system dynamics,
and metal ecology. J. Ind. Ecol., 23–43.
Webster, B.Y., 2011. Understanding & Comparing Risk. URL ⟨http://reliabilityweb.
com/index.php/print/understanding_and_comparing_risk⟩.
Wilson, A., McMahon, F., Cervantes, M., 2013. Global Mining Survey—Results for
2012/2013: Policy Potential Index 2013. Fraser Institute, Vancouver, BC, Canada.
Yale Center for Environmental Law and Policy, 2014. Environmental Performance
Index—Summary for Policy Makers. Technical Report. Yale University.