Figge et al 2018.pdf

Ecological Economics 150 (2018) 297–306 Contents lists available at ScienceDirect Ecological Economics journal homepage: www.elsevier.com/locate/ecolecon Analysis Longevity and Circularity as Indicators of Eco-Efficient Resource Use in the Circular Economy T Frank Figge , Andrea Stevenson Thorpe, Philippe Givry, Louise Canning, Elizabeth Franklin-Johnson ⁎ Kedge Business School, Domaine de Luminy – BP 921, Rue Antoine Bourdelle, 13288 Marseille CEDEX 9, France A B S T R A C T Natural resources are limited. The circular economy is one of several different concepts that has been useful in the quest to understand how resources can be used most efficiently. It proposes that closing loops and repeatedly using resources has the potential to procure maximum eco-efficiency. To track society's progress towards a circular economy, indicators and measures are needed. The majority of these aim to capture the circularity of resource flows, yet fail to simultaneously consider the length of time for which a resource is in use. More recently, a longevity indicator has been proposed, but similarly, it fails to take into account how many times a resource is used. Both longevity and circularity are needed for sustainable resource use, but to date, no measure that combines both approaches is in use. Based on existing measures we develop and further develop indicators for both circularity and longevity that focus on the contribution that organisations and other resource users make to the sustainability of resource use. By combining both indicators we enhance their explanatory power. 1. Introduction Natural resources on planet earth are limited. The rate at which we use these has long been noted as unsustainable (Behrens et al., 2007; Green, 1894; Malthus, 1798; Meadows et al., 1972). As society moves towards strategies of sustainable development, debates have emerged on how resources can be used more efficiently (Daly, 1990). One proposition has been the concept of a circular economy (Pearce and Turner, 1990). Based on seminal work by Leontief (1991), the underpinning idea is to repeatedly use the same resources in a loop, decoupling precious stocks of virgin resources from economic activity. In this approach to sustainability, the way in which firms choose to use resources becomes key: More sustainable living at the societal level can only occur when organisations use resources more efficiently. A plethora of metrics have emerged that aim to capture the extent of firms' contributions to a circular economy. Common to most indicators is the assumption that the efficiency of a resource can be measured by the number of times it is used, i.e. its circularity. Conversely, FranklinJohnson et al. (2016) propose a different approach. They emphasise the length of time a resource is in use, i.e. its longevity. In this article, we start with the premise that both circularity and longevity based measures have their merits. In other words, we recognise the value of both approaches - the frequency and length of time a resource is used - to determine resource efficiency. However, we also note that, in practice, a resource's circularity is not necessarily an indicator of its longevity. A resource could be used many times, but within a short time span, for example. We also argue that a resource's longevity is not necessarily an indicator of its circularity. A resource could be used very few times, but over a long time period. Put simply, if both measures were applied independently to assess how efficiently firms use resources, the two measures could very feasibly draw two different conclusions. We propose value in combining both approaches, in contrast to existing measures that have a singular focus on either circularity or longevity. The two-dimensional indicator that we propose, enhances the individual strengths of circularity and longevity, whilst avoiding the problem of incompleteness, as outlined above. Our approach also enables different organisationally relevant strategies to increase resource use efficiency in practice. In particular, the scope of our indicators extends to the longevity and circularity of resources within a given product system. It is this product system that defines the boundaries of our indicators, rather than the resource itself that could be recycled, for example, beyond the product system. We structure our article as follows. In Section 2, we discuss circularity and longevity as drivers of eco-efficiency. In Section 3, we develop a measure of circularity, followed by a measure of longevity. In Corresponding author. E-mail addresses: figge@sustainablevalue.com (F. Figge), Andrea.thorpe@kedgebs.com (A.S. Thorpe), Philippe.givry@kedgebs.com (P. Givry), Louise.canning@kedgebs.com (L. Canning), Elizabeth.franklin-johnson@kedgebs.com (E. Franklin-Johnson). ⁎ https://doi.org/10.1016/j.ecolecon.2018.04.030 Received 3 November 2017; Received in revised form 23 April 2018; Accepted 24 April 2018 0921-8009/ © 2018 Elsevier B.V. All rights reserved. Ecological Economics 150 (2018) 297–306 F. Figge et al. hierarchical arrangement of reuse, repair, refurbishment, remanufacturing, repurpose, and finally recycling (Bocken et al., 2017). Most high-level policy (e.g. European Commission) is congruent in favouring reuse over recycling in their ‘waste hierarchies’. Importantly though, firms in practice tend to favour recycling over refurbishment (Allwood, 2014; Potting et al., 2017). We now explore some of the various circularity approaches and metrics in more detail, along with the concept of longevity that challenges the completeness of circularity as an eco-efficiency measure. Section 4, we combine circularity and longevity approaches, and describe a case study that showcases the practical applicability of our measure. We discuss our findings in Section 5, before concluding in Section 6. 2. Eco-Efficiency and the Circular Economy A growing and increasingly affluent population demands more resources to fulfil its economic needs and desires (Krausmann et al., 2009; Pimentel and Pimentel, 2006; Wiedmann et al., 2015; York et al., 2003). The problem is that on “spaceship earth” (Boulding, 1973, 1996) natural resources are finite. Yet, economic activity, not least as one crucial part of sustainable development – is juxtaposed with environmental and social concerns (Elkington, 1997). The challenge, therefore, is to address the tension between dwindling natural resources, and the need for greater economic activity (Commoner, 1972). “Eco-efficiency” (World Business Council for Sustainable Development, 1996, 2000) goes some way to meet this challenge (Caiado et al., 2017; Charmondusit et al., 2014; Hoffrén and Apajalahti, 2009; Huppes and Ishikawa, 2009; Müller et al., 2015). The underpinning idea is that less natural resources are used to generate the same, or a greater, amount of economic activity. The concept is well-established in both theory and practice, and across a wide variety of contexts. Charmondusit (2017), for example, shows how the design process of kitchen furniture enhances eco-efficiency in a product setting. Peng et al. (2017) describe the concept within a service environment - in their case, tourism (see also Strasburg and Jahno, 2017; Zhang et al., 2014). Others have used eco-efficiency principles to drive the design of manufacturing plants (May et al., 2016) and industrial parks (Grant, 1997), or to inform policy (e.g. World Business Council for Sustainable Development, 1996). Whatever the setting, objectives are loosely grouped around sustainability. In short, eco-efficiency is a tool for sustainability analysis and development (Huppes and Ishikawa, 2005; Maxime et al., 2006; Zhang et al., 2008). Furthermore, it potentially enables ‘better’ decisionmaking, both in practice and at a policy level, on the way in which natural resources are used (Van Caneghem et al., 2010). Lozano and Lozano (2017) for instance, describe a case study in which managers use eco-efficiency to choose between two different processes to transform residual biomass into chemicals or energy as outputs (see also, Alves and de Medeiros, 2015; Carvalho et al., 2017). Decision-making, however, requires metrics and the need for ecoefficiency measures has never been greater (Van Caneghem et al., 2010). One popular conceptual approach which has emerged is the pairing of eco-efficiency with the circular economy (Pearce and Turner, 1990). Although fragmented (Korhonen et al., 2018), most circular economy scholars agree on its high level meaning: By closing loops and making resource use circular, resources are not lost, but are instead used repeatedly. This has the effect of increasing eco-efficiency (Yuan et al., 2006). From this premise, a rich, albeit differentiated, array of metrics that measure ‘circularity’ has emerged (Blomsma and Brennan, 2017; Bocken et al., 2017; Pauliuk, 2018). In particular, the circularity of virgin resources is important (e.g. Figge et al., 2017). Such resources are increasingly scarce (Boulding, 1973, 1996), but a circular use decreases the pressure on the finite amounts that are hypothetically still available. Thus, circularity metrics encourage organisations (Lewandowski, 2016), sectors (Parajuly and Wenzel, 2017), industries (Cayzer et al., 2017), regions (Smol et al., 2017) or policy makers (EU Commission, 2015) to use virgin resources more circularly, i.e. to increase their eco-efficiency. Key for organisations and policy-makers therefore, is to understand the way in which the combined concepts of eco-efficiency and circularity translate into tangible activities. This has encouraged some researchers to focus on identifying a hierarchy of preferred tactics, i.e. those which maximise the extent of circularity. The ‘butterfly figure’ from Ellen MacArthur Foundation, for instance, has influenced a 3. Measuring Resource Use As independent measures, both circularity and longevity indicate the extent to which virgin resources are used eco-efficiently. In theory, if the circularity of resource use is infinite, its longevity is also infinite. However, in practice this does not occur, and there is usually a greater degree of separation between the two. We discuss circularity first. 3.1. Circularity In the circular economy, nutrients and material resources circulate and remain within biospheres and product systems (United Nations Environment Programme, 2012). Understanding the flow - or circularity - of such resources requires a metric, and preferably one that captures resource use at the level of the individual organisation. Measures have been proposed by numerous scholars, who have also pointed to the limitations of such measures (see Pauliuk, 2018, for instance, for a recent overview). Some measures focus on sectors and products and draw from material flow analysis to do this (e.g. Lai et al., 2008; Wen and Meng, 2015). They typically examine supply chains, to determine the inputs and outputs of specific resources (and their combination), from their initial primary extraction to their eventual disposal as waste. In theory, monitoring the flow of resources enables firms to adopt practices that could maximise their use of materials derived from recovery and recycling, whilst minimising the use of virgin resources. A limitation of this approach is that in reality, material flow analyses tend to be more useful at industry, national and international levels, rather than for individual firms. A second limitation is that using more reused/recycled materials and less virgin resources does not necessarily equate to greater circularity. The former can originate from other products, for instance, and some sectors have fully linear flows. Although this type of measure can establish circularity per se, i.e. that resources are retained within product systems, it fails to capture the number of times they pass through product systems. Recycling and reusing resources are necessary for circularity, but are insufficient as a measure on their own. Bailey et al. (2004) partially address this issue by suggesting that circularity can be measured via ‘path length’. This is defined as the average number of processes in which resources are involved, before they leave the product system as waste (Bailey et al., 2004). In contrast to general material flow analyses, as above, this metric is intended for individual firms. Bailey et al. (2004), for instance, examined the value chain of a carpet manufacturer, showing how path length is extended through repeated cycles of production, consumption, material separation, and reclamation. The emphasis on process, however, generates its own limitations. First, path length will depend on how system boundaries and processes are scoped. Furthermore, the degree of complexity and connectivity of processes that involve a high division of labour will also affect conclusions of path length. In short, measuring circularity in this way will not necessarily reflect the eco-efficiency of resource use. Second, one of the common problems with which circularity metrics grapple, is that not all materials or products can be repeatedly used for the same purpose. It is only in closed material systems that a material is recycled into the same use with unchanged material properties 298 Ecological Economics 150 (2018) 297–306 F. Figge et al. implies circularity according to the proportion of recirculated content, it does not quantify the number of times that material resources or parts pass through different phases in a value chain. Third, the practical challenge for some organisations to calculate the costs of recirculated materials within complex products, especially when data is required from other locations in the value chain, might be too great. (Dubreuil et al., 2010; Graedel et al., 2011). Aluminium cans that are recycled to become aluminium cans once again, is an example of a closed material system (Dubreuil et al., 2010). In open material systems, a material is recycled into different product systems, and the properties of the material change. The tin coating inside steel cans that is mixed with steel when they are recycled is an example of an open material system (Dubreuil et al., 2010). It is only in the former system that the primary use of resources is displaced (Graedel et al., 2011). As argued by Geyer et al. (2016) “it matters not only how much material is recycled, but also what it is recycled into and therefore can displace” (p. 1011). A truly circular use of resources can only be achieved within closed material systems. In open material systems, resources can be used multiple times, but in ways that are different to their original use. In open material systems, therefore, recycling is unable to displace materials in the original process, resulting in a low level of circularity. Circularity has been developed most actively at the sector level, and across regional and global geographies (e.g. Daigo et al., 2005; Eckelman and Daigo, 2008; Pauliuk et al., 2017; Pauliuk et al., 2011). Daigo et al. (2005) describe how iron is used around 2.67 times on average in society, for example. Similar studies calculate a circularity of 3 for nickel (Eckelman et al., 2012), and 1.9 times for copper (Eckelman and Daigo, 2008). These research endeavours focus on the use of a resource across a range of processes, and address the question of how often resources are used before being lost. Bailey et al. (2008) offer an adaptation for individual firms. They propose a simplified metric for circularity and, for illustrative purposes, apply this to the product category of tyres. Although their indicators such as ‘retreading’ are highly specific, they can still be abstracted and applied to alternative products. However, their choice of product may be too simple to generate any meaningful extrapolation of circularity indicators to more complex products. Further development of indicators, especially those which capture the performance of multiple material resources within complex products, has not been forthcoming. Ellen MacArthur Foundation and Granta Design (2015), and Linder et al. (2017), also note their absence, and propose their own measures to address this. The former describe a Material Circularity Indicator (MIC) that consists of ‘flow’ and ‘utility’. Flow reflects the principles of material flow analysis. Its linear flow index measures the proportion of virgin material that follows a path, thus creating unrecoverable waste. Alongside flow, a utility factor is proposed, combining the time for which a product is used (lifetime) and the extent to which a product is used to its full capacity (intensity). The authors themselves concede that, in practice, only part of the utility factor would be used, owing to the complexity of the overall metric. A more fundamental limitation stems from each component of their measure relying on estimated figures, rather than verifiable data (Linder et al., 2017). Additionally, when amalgamating the scores for the different material resources within products, it is difficult to accommodate the variety of reuse and recovery rates for each resource (Ellen MacArthur Foundation and Granta Design, 2015; Linder et al., 2017. Linder et al. (2017) base their approach on economic value. They calculate circularity according to the value of recirculated parts within a product, as a proportion of its total constituent parts. To measure economic value, Linder et al. (2017) use cost to the product vendor. The rationale is that whilst reconditioned parts and recovered materials will incur different processing costs, these can be combined into a single figure. As a result, their cost based indicator allows for the amalgamation of different materials, components, and processing actions. It can be used by different vendors within a value chain, and it enables comparisons between various products. Further, it could potentially be used to track the performance of resources, and thus inform decisionmaking. We do however see some flaws in this. First, as Linder et al. (2017) themselves acknowledge, the indicator fails to capture how long product systems retain resource materials. Second, whilst the measure 3.1.1. The Circularity Metric We propose a metric that addresses some of the limitations of the circularity measures that we have discussed. With these in mind, our overall objective is to develop a tool that can capture the contributions that organisations make to the circular use of resources. To be relevant to firms, our metric purposefully only extends to activities which the firm can control, in contrast to some other metrics and approaches (e.g. Ellen MacArthur Foundation and Granta Design, 2015; Linder et al., 2017). We also scope the resource within the product system, rather than focusing on the resource in a wider context, which might otherwise be recycled, for example, into alternative products outside the boundaries of the firm. Further, we design our measure in a way that accommodates the primary function of companies, i.e. their raison d'être is to produce goods or services that satisfy human needs. We see processes only as a means to an end, rather than the purpose of firms per se. These features reveal four important differences between our metric and previous indicators. First, we locate ‘resource use’ as being within the goods or services that are produced by firms, rather than in the context of processes. We are therefore interested in the amount of resources embodied in a product or service rather than the efficiency of the processes that have produced them. This contrasts with the assumptions of most material flow analyses, for example. Second, we scope ‘resource use’ only within the firm. This contrasts with the cradle-to-cradle (Braungart et al., 2007; McDonough and Braungart, 2010) and cradle-to-grave approaches of many life cycle assessments. Realistically, firms might influence their supply chains only weakly at best. Our metric only extends to the resources of which firms have control, and how they keep these within the product system, i.e. use, reuse (or refurbishment), and recycling. We propose that this generates a more valuable tool for managers. Third, we focus our metric as a whole on circularity, i.e. the number of times a resource is used. Thus, ‘physical’ information drives our metric, rather than any financial or monetary data that many ecoefficiency measures use. Adding a valuation of inputs (e.g. their cost) or outputs (e.g. the utility of products) might be insightful. However, we argue these should be separated from circularity assessments, as they essentially have different explanatory purposes. Fourth, as we stay within a scope that can be controlled by a decision maker. The model that underlies our indicator is therefore deterministic. 3.1.2. Calculations We identify three different contributions to circularity that we depict in Eq. (1). These are: ‘initial use’ (NA); ‘refurbishment’ (NB); ‘recycling’ (NC). Thus: Circularity = N A + N B + N C (1) Circularity is expressed as the number of times a resource is used in a product system. The three components (NA, NB, NC) are therefore analogously expressed as the contribution to the number of times a resource is used in a product system. In the following we first develop all three contributions separately before we combine them to the circularity metric. Each initial use means that material is used once, as we depict in Eq. (2). Thus: NA = 1 (2) B To determine the contribution of refurbishment (N ), we calculate the proportion of the initial resource that is reused. This corresponds to 299 Ecological Economics 150 (2018) 297–306 F. Figge et al. A B C ⎧Circularity = N + N + N ⎪ NA = 1 ⎪ 1 − (ab)n N B = (ab) ⎡ 1 − (ab) ⎤ ⎨ ⎣ ⎦ ⎪ p 1 − (ab)n + 1 ⎪ N C = 1 − p ⎡ 1 − (ab) ⎤ ⎣ ⎦ ⎩ the sum of the products of the goods returned (aj), and the proportion of those that are returned, which are then refurbished (bj). We depict this in Eq. (3). The variable n shows the total number of cycles there are. For each cycle i the amount of material that passes through each cycle preceding and including cycle i must be considered. This is done by multiplying the percentages of products that are returned (a) and refurbished (b) for all cycles preceding and including cycle i. The sum of these products shows how much material is refurbished. Thus: NB = ⎡⎛ n i=1 i ⎢ ⎣ ⎝ j=1 The first term of N corresponds to a geometric series. This geometric series considers the theoretically infinite use of resources that are recycled and re-enter the system to be used in new products and services. When recycling is not perfect, but resources are being wasted, an ever-decreasing fraction of the resources will re-enter with each p cycle. The term 1 − p shows the percentage of resources that are recycled overall after an infinite number of cycles have been taken into account. This circularity measure reflects the amount of times a resource is used on average in a product system. Circularity can be between 1 and infinity. A circularity of 1 means a fully linear system; a circularity of infinity means a fully circular product system. ⎞⎤ ∑ ⎢ ⎜∏ aj bj⎟ ⎥ ⎠⎥ ⎦ (3) So, for example, if overall 50% of goods are returned, and 50% of these returned goods are then refurbished, then 25% of all the materials within these goods are reused after the end of the first application of the material. To determine the contribution of recycling (NC), we assume a process that starts with the return of goods (aj), which are then dismantled and recycled (ci in the ith additional cycle). The proportion of the overall resources that can be recovered from recycling also depends on the percentage of the resource that can be recovered from each product (di in the ith additional cycle). These resources then enter the process again. A proportion of these products will be recycled again, and we use an iterative process that depends on the proportion of the initial material that is recycled (p, as defined in Eq. (4a)). The variable p reflects the fraction of the initial material that will be recovered through recycling. Some of the material will be recovered from recycling products after their initial use. This corresponds to the first term (a1c1d1) of Eq. 3.2. Longevity An alternative way to measure the eco-efficiency of virgin resources is to emphasise longevity, i.e. the length of time that a resource is used, which can be measured in days, months, years, and so on (FranklinJohnson et al., 2016). The idea is that the longer a resource is used, the higher the contribution to a circular economy. High circularity is still important, but only to the degree to which it contributes to a longer time period that the resource remains ‘active’. The specific number of times as an isolated measure is unimportant. In this sense, circularity is a means to an end, rather than an end itself. Longevity is determined in three ways: the time for which a resource is first used (A); the time for which a resource is used due to product refurbishment (B), and due to recycling (C). In their measure, FranklinJohnson et al. (2016) use an example of a process where goods are refurbished and resources are recycled, each to a maximum of twice. They describe longevity as the sum of three contributions, as depicted in Eq. (6): The initial lifetime of the product (LA); refurbished lifetime contribution (LB); recycled lifetime contribution (LC). Thus: (4a). The second term of Eq. (4a) (∑i = 2 ⎡ ∏ j = 1 aj bj ai ci di⎤) reflects the ⎣ ⎦ material that is recovered from products that have been refurbished once or more often and that are subsequently recycled. n p = a1 c1 d1 + ⎡ ⎛ i−1 ⎞ ( i−1 ) ⎤ ∑ ⎢ ⎜∏ aj bj ⎟ ai ci di⎥ n i=2 The ratio of ⎢ ⎣ ⎝ j=1 p 1−p ⎠ ⎥ ⎦ (4a) shows the percentage of material that is recycled after taking into account that some of the material will be recycled several times. This recycled material will enter the product system from the start and will make yet again thee contributions to circularity: Through ‘initial use’ (NA), ‘refurbishment’ (NB) and ‘recycling’ (NC) of products. We depict the overall contribution of recycling in Eq. (4b). Thus: N C = p (N A + N B + N C ) ⇔ N C = p (N A + N B ) 1−p Longevity = LA + LB + LC c=1−b (6) Longevity is measured in units of time (f. ex. years or months). The three components (LA, LB, LC) are therefore analogously expressed as the contribution to the length of times a resource is used in a product system. In one sense, the work of Franklin-Johnson et al. (2016) significantly advances understanding of the measurement of resource ecoefficiency by offering an alternative approach that addresses many of the limitations of circularity measures. However, whilst Franklin-Johnson et al. (2016) calculate longevity for a particular case, their model fails to accommodate different frequencies of return, refurbishment, and recycling. Thus, whilst their rationale and calculations for longevity are sound, their application is somewhat limited. Here, we adapt their original formula to reflect these criticisms. (4b) We now refer to our previous example where 50% of goods are returned. Of these returned goods, half enter into the recycling process, i.e. 25% of the overall goods sold are recycled. If this process is able to recover 50% of the material within recycled goods, then, overall, 12.5% of resources can be recovered through recycling. This corresponds to a value of p of 12.5%. To calculate the contribution of recycling (NC) we need to take into account that the recycled resources will make a fractional contribution to another initial use (NA) and refurbished use (NB). This is what Eq. (4b) calculates. As NA corresponds to 1 and NB to 0.25 in our example we can calculate a value of NC of 0.18. In total there is therefore a contribution of 1.43 to the circularity of resource use. With two realistic assumptions we can generate a more manageable and informative formula. These are that: return, reuse, recycling, and recovery rates are constant (A1); all returned products are either reused or recycled (A2). Thus: ai = a, bi = b, ci = c, and di = d whatever i (i = 1,…, n) (5) C 3.2.1. Calculations We refer to the same parameters that Franklin-Johnson et al. (2016) use to scope longevity, i.e. Eq. (6), as above. LA and LB take into account the potentially different lifetimes of recycled and reused products. The initial lifetime contribution (LA) contributes to the amount of time during which a product is used initially, i.e. before it is discarded, (A1) (A2) Circularity, and its intermediate components, therefore becomes1 1 300 See Figge et al. (2017), Appendix B for the spirit of proofs. Ecological Economics 150 (2018) 297–306 F. Figge et al. this example, we arrive at a contribution to longevity of 1.93 months. In total longevity amounts to 15.43 months. By making three realistic assumptions, we can generate a more manageable and informative formula. The first two mirror the assumptions that we made in our circularity metric, depicted in Eqs. (A1) and (A2). The third assumption is that the lifetime of a product through refurbishment decreases constantly with every step. This is depicted in Eq. (A3). Thus: refurbished and reused or recycled. To determine the refurbished lifetime contribution (LB) all possibilities that could lead to this are acknowledged: The time periods (expressed in the same relevant unit as for the initial use) that a product is used in an ith cycle (LiB), are multiplied by the proportion of the initial resources that are eventually reused. This corresponds to the product of returned items (aj) and the proportion of those products that are refurbished (bj) after the first initial use and until the nth additional cycle. Eq. (7) depicts this. Thus: LB = ⎡⎛ ⎞ ⎤ ∑ ⎢ ⎜∏ aj bj ⎟ LiB⎥ n i=1 i ⎢ ⎣ ⎝ j=1 ⎠ LiB = αLiB− 1, whatever i (i = 1,…, n), with L0B = LA In conclusion, these three assumptions enable us to rewrite Eqs. (7) and (8a), as (the terms in square brackets representing the contributions of refurbishment and recycling, respectively, for increasing longevity): (7) Recycled lifetime (L ) is depicted by Eq. (8a). This corresponds to the lifetime that is created by the process of returning products (ai in the ith additional cycle), dismantling them, and recycling the resources (ci in the ith additional cycle). The proportion of the overall resources that can be recovered through recycling also depends on the percentage of the resource that can be recovered from each product (di in the ith additional cycle). This reflects the usual impossibility of being able to recover 100% of resources within products.2 These resources are used again, and, in turn, create an additional initial lifetime (LA) as new products are created, and a refurbished lifetime (LB), which can be different. A fraction of these products will be recycled once again in the same product system. To address this infinite cycle, Franklin-Johnson et al. (2016) use an iterative process that depends on the percentage of material recovered through recycling (p in Eq. (8b)). The variable p reflects, similarly to Eq. (4a), the fraction of the initial material that will be recovered through recycling. It ranges from 0% (no recycling) to 100% (all materials are fully recycled). LC = p (LA + LB + LC ) ⇔ LC = p (LA + LB ) 1−p A B C ⎧ Longevity = L + L + L ⎪ B n ⎪ L = LA ⎡ (ab α ) 1 − (abα ) ⎤ 1 − (abα ) ⎦ ⎣ ⎨ 1 − (abα )n + 1 ⎪ LC = LA ⎡ p ⎤ ⎪ 1 − (abα ) ⎣ 1−p ⎦ ⎩ ( ( )( ⎡ ⎛ i−1 ⎞ i=2 ⎢ ⎣ ⎝ j=1 ⎠ ⎥ ⎦ ) (9) Both measures are limited in the extent to which they are able to express resource use as a contribution to the circular economy. Whilst circularity measures show the average number of times a resource is used, they indicate nothing about the period of time for which they are used. Feasibly, resources could be used many times, but only for a short duration overall. Conversely, resources might be used for a long period of time, but not in a circular way - something that a longevity approach by itself fails to address. We address these limitations by combining both approaches. In doing so we propose a measure that has greater explanatory power than that offered by circularity and longevity measures individually. Our Combination Matrix (Fig. 1), identifies four possible ways in which circularity and longevity could be combined. The matrix represents a simple framework of resource use. On the one hand, firms could use this to identify how they currently use resources, and to then develop strategies for more sustainable usage. Keeping in mind the advantages proposed by circularity and longevity as individual measures, actions that move the firm towards the most sustainable use of resources, i.e. a ‘long circular’ use, are the most desirable as they use resources to their fullest potential. On the other hand, our matrix can be improved by a more sophisticated analysis of resource use. By distinguishing between initial use (A), refurbishment (B), and recycling (C), and by applying these components to our matrix, we can more adequately illustrate just how firms can begin to use resources in a ‘long circular’ way, i.e. more sustainably. Combination Matrix B in Fig. 2 demonstrates a firm that might initially identify its use of resources as ‘short linear’. By refurbishing and reusing products, the firm moves from point A to B, and in doing so starts to use its resources in a ‘long linear’ way. Then, adopting recycling practices, the firm is able to move from point B to C, i.e. the firm starts to use its resources in a ‘long circular’ way. Importantly, the length and the direction of the arrows in Combination Matrix B signify an increasingly more sustainable use of resources. The more horizontal the arrows are in Combination Matrix B, the higher the contribution to longevity, in relation to the contribution to circularity. At the same time, the specific path - signified by the length and gradient of the arrows in particular - is specific to the type of resource (and the firm) under assessment. For example, some resources, and the products in which they are embodied, have extraordinarily long initial use periods. To demonstrate our measure, we use the same example as FranklinJohnson et al. (2016) i.e. mobile handsets as the product, and gold as the resource. We calculate the circularity according to Eq. (5) and (8a) ⎤ ∑ ⎢ ⎜∏ aj bj ⎟ ai ci di⎥ ) 4. Combining Circularity and Longevity with n LiB = α iLA (A3) ⎥ ⎦ C p = a1 c1 d1 + ⇔ (8b) Longevity describes the time that a resource is used. A longer time period of use makes a higher contribution to sustainable resource use than a shorter period. We can return to the example above to show how the three components of longevity (LA, LB and LC) are calculated. We assume once again that 50% of goods are returned. Of these returned goods, half enter into the recycling process, i.e. 25% of the overall goods sold are recycled. The other 25% are refurbished and reused. The process recovers 50% of the material within recycled goods, i.e. overall, 12.5% of resources are recovered through recycling. This corresponds yet again to a value of p of 12.5%. As above we calculate the contribution of recycling (LC) by looking at the fractional contribution to another initial use (LA) and refurbished use (LB). For this we need to know for how long a product is used initially (LA). We assume that this corresponds to 12 months. Furthermore, we need to know for how long refurbished products are used. We assume that refurbished products are always used for 6 months. Longevity is the sum of the three components LA, LB and LC (Eq. (6)). LA, the initial use time, corresponds to 12 months as just defined. We know that 25% of all products are refurbished. Given that refurbished products are used for six months this adds another 1.5 months to longevity (LB). The recycled materials will create additional longevity by creating new products of which some will be refurbished. Some of the materials will also be recycled once again. This additional lifetime is described by LC (Eq. (8a/8b)). For 2 On this point, Franklin-Johnson et al. (2016) are erroneous. They use the amount of unrecoverable material, rather than the amount of recoverable material in their formula. We address this error here. 301 Ecological Economics 150 (2018) 297–306 F. Figge et al. Fig. 1. Combination Matrix. Fig. 2. Combination Matrix B and resource use strategies. Table 1 Summary of data and sources used in the applied model (based on: Franklin-Johnson et al., 2016). Description Keya Figure Source Keyb Handset's initial lifetime Handset's second-hand lifetime A U1 24 months 12 months Geyer and Blass (2010); Paiano et al. (2013); Wilhelm et al. (2011) Industry experts estimate second-hand use to be half the initial lifetime due to degeneration and technological obsolescence Handsets outside market control Handsets returned Handsets repaired or refurbished and resold Handsets recycled Precious material retained through recycling Number of additional cycles v1 w1 x1 85% 15% 65% Ellen MacArthur Foundation (2012) Ellen MacArthur Foundation (2012) Geyer and Blass (2010) LA(=L0B) L1B hence, ∝ = 50% (1 − a) a b y1 z1 35% 95% Geyer and Blass (2010) Hagelüken (2007); Rochat et al. (2007) c = (1 − b) d a b 2 n Name of variables in Franklin-Johnson et al. (2016). Corresponding variables in this article. 302 Ecological Economics 150 (2018) 297–306 F. Figge et al. 5. Discussion Table 2 Longevity and circularity of gold in mobile phones. Initial use (A) Refurbishment (B) Contribution of refurbishment (% points) Recycling (C) Contribution of recycling (% points) Longevity (=26.69) Circularity (=1.17) LA LB LB/LA 24 months 1.23 +5.12% NA NB NB/NA 1 0.11 +10.7% LC LC/LA 1.41 +5.87% NC NC/NA 0.061 +6.06% Circularity and longevity are two approaches for assessing the sustainability of resource use. We have discussed the strengths and weaknesses of each, and these motivated the development of our own measure. Succinctly, in this article we have described an approach that merges principles of circularity and longevity. In this section, we first briefly revisit the correction we made to the calculations presented by Franklin-Johnson et al. (2016), as a minor contribution of our article. We then discuss our approach in more detail, particularly in terms of how it contributes to the conceptual basis of resource use measurement. We then describe the contributions of our own metric, as in Section 4. Both our conceptual approach and its related measure constitute this article's principle contribution. We finish by discussing the implications of our work for policy and practice. We finally conclude with some limitations and suggestions for further research. longevity according to Eq. (9) and we depict the results graphically. Table 1 provides the data for the use of gold in mobile handsets. We correct Franklin-Johnson et al. (2016) error on the recovery rate of materials in recycling, as above. Table 2 presents the results of the calculation of longevity and circularity for this example. We can now illustrate the results from our calculations by using our Combination Matrix as a basis. Fig. 3 depicts the combination of longevity and circularity as relating to handsets and gold. Fig. 3 indicates three points, but with emphasis on the activity that connects each point to the next. Point A indicates that the resource, i.e. gold, has been used once, for a duration of 24 months. Refurbishment of handsets increases the circularity of the resource by 10.7%. It also extends the time period for which the resource is used, i.e. its longevity, by 5.12%. This brings us to point A + B, i.e. the resource's initial use and refurbished use. From here, we can see that recycling further increases the circularity of the gold (by 6.06%), as well extending the period of time for which it is used (by 5.87%). It is interesting to note that refurbishment adds more to circularity than recycling. In the case of longevity the opposite holds true: recycling contributes more than refurbishment. Overall, point A + B + C represents resource use as a sum of both refurbishment and recycling activities. This point represents a total of almost 17% that is added to the circularity of the gold within handsets, and about 11% to its longevity, from refurbishment and recycling. We can conclude therefore, that for this particular resource, refurbishment and recycling add more to the circularity of gold, than to its longevity. 5.1. Correction to Franklin-Johnson et al. (2016) The work from Franklin-Johnson et al. (2016) presents an interesting and stimulating challenge to the more common approach (to date) of circularity. ‘Time’ as a basis for measuring resource use, had not been addressed by research in a way that centralises its role in assessing sustainability. In this article we correct an error of FranklinJohnson et al. (2016), where they mistakenly incorporate the amount of unrecoverable material, rather than the amount of recoverable material into their formula. We address this error in our own calculations in Section 3.2.1. We see this as a minor contribution of our article, yet it is important for the implications that it has for future work. Franklin-Johnson et al.'s (2016) work represents the first in longevity measures, thus, we assume that any future research taking this approach will refer to their article, and either reference or use their formula. Therefore, we argue that correcting their error is of prime importance for the development of a longevity approach in itself, as well as for the advancement of the combination metric that we propose in this article. 5.2. Our Conceptual Approach The combination of circularity and longevity equates to a different conceptual approach than previously established. Our approach is far 1.18 A+B+C +6.06% 1.16 1.14 A+B 1.1 1.08 +10.7% Circularity [nb] 1.12 1.06 1.04 1.02 1 A +5.12% +5.87% 0.98 23.5 24 24.5 25 25.5 26 26.5 Longevity [months] Fig. 3. Circularity and longevity of gold in mobile phones using the Combination Matrix. 303 27 Ecological Economics 150 (2018) 297–306 F. Figge et al. component. from a simple sum of the two individual approaches. Instead, our analysis has yielded a wholly different approach, but which takes into account the major strengths and weaknesses of both. In the hypothetical scenario of a perfect circular economy, resources would be used an infinite number of times. In addition, they would be used for an infinite amount of time. With no wastage, both circularity and longevity approaches arrive at the same conclusion, i.e. that resources are maximally sustainable. In reality, however, resources are only used for a finite number of times - and for a limited amount of time - before being designated as waste. It is these more realistic scenarios in which the distinction between the two measures becomes important. An example of beverage cans illustrates the point. The aluminium within cans is reused multiple times, but is eventually discarded as waste after only a relatively short period of time. A circularity approach would conclude the aluminium to have made a high contribution to sustainable resource use. A longevity approach though would argue otherwise, pointing to the short period of time that the aluminium stays within the product system. A second example, indicates the converse. The stone of the ancient Parthenon in Athens has remained within the same initial use for over 2000 years. A longevity approach would conclude a sustainable use of the stone; a circularity perspective would say otherwise. Such a distinction, essentially calls into question the reliability of both approaches when considered individually. Combining both into one concept seems prudent, especially when it exploits the strengths of both approaches. The underpinning logic behind each approach in its very broadest interpretation, for example, is clear. It seems to make sense, for instance, to assess the eco-efficiency of resources by looking at how long they are used for. When we explore circularity and longevity more deeply, further limitations of previous conceptualisations become apparent and we address these. First, our concept is easily applicable at the organisational level. Some versions of circularity are best suited to industry or other macro environments - those which focus on supply chains, for example. That said, our concept has a high degree of flexibility. Whilst we have scoped our concept primarily with the organisation in mind, there is no reason as to why our concept cannot be applied to industry, or even national or international contexts. Thus, we propose a conceptual approach that is not only truly relevant to organisations, but which can also be applied to macro environments. This is in contrast to many of the interpretations of circularity, in particular, which tend to focus on one level or the other. Second, we argue that our approach examines resource use more objectively than other approaches that focus on processes. By locating resource use within products themselves, i.e. not within their underpinning processes, there can be little variation in how the domain of such resource use is defined. In contrast, determining the scope of a process is often arbitrary and subjective. When critical components of a process are outsourced, for example, they may or may not be included in assessments, potentially leading to differentiated conclusions on whether resources are used efficiently or not (Behrens et al., 2007). Our approach avoids these particular issues of subjectivity by focusing on products or services. In doing so, we suggest our approach is more objective in the way that it determines the location of resource use. Third, we strip back the concept of circularity to its literal meaning, i.e. we start with the notion of circularity as something which describes how many times a resource stays within a product system. As we have argued, recycling and reusing products, for example, may be part of a process of circularity, but do not represent circularity itself. Thus, we propose a conceptual approach that refers to circularity in its truest sense. Finally, our concept takes longevity and extends the notion to a more sophisticated level. In particular, we introduce the ideas of ‘reuse’, ‘recycling’, and ‘refurbishment’, something that current longevity work fails to do systematically. Thus, we propose a conceptual approach that marries well-established concepts of sustainability to its longevity 5.3. Our Combination Metric Our indicator makes a contribution as a powerfully reliable metric of resource use assessment. Derived from our conceptual approach, we describe a method of calculating resource use, particularly, but not exclusively, at firm level. We use a corrected and generalised version of Franklin-Johnson et al.'s (2016) formula for longevity. We combine this with circularity to create one metric that is more than the mere sum of its parts. We adapt each concept into a single formula by addressing some of the weaknesses of each approach, whilst taking note of their strengths. However, the limitations of circularity are addressed by the strengths of longevity - and vice versa. A combined metric that unites principles of both circularity and longevity has not been proposed before. In doing so, we argue that our metric has a higher degree of reliability than circularity and longevity offer as individual measures. Further characteristics of our metric add to its overall appeal and contribution to the measurement of sustainable resource use. Thus, in addition to its reliability, we are able to describe our metric as: organisationally relevant, insightful, and user-friendly, applicable to multiple resources and products with varied extents of complexity, and driven by physical data. This ensures a higher degree of validity than previous work, as we go on to argue. Circularity measures have thus far failed to produce a metric that strikes a balance between being organisationally relevant and sufficiently sophisticated to capture resource use within complex products. Down-cycling approaches, for example, are better suited to larger macro scenarios than those pertaining to individual firms. The formula proposed by Bailey et al. (2008) can arguably only capture resource use within very simple products. We argue that our measure is not only organizationally relevant, but is flexible enough to assess resource use within goods and services of varied complexity. Thus, we propose a metric that can be used by firms to assess a multitude of different products. Some circularity measures have been developed to capture a much wider range of complex products with multiple resources (e.g. Ellen MacArthur Foundation and Granta Design, 2015; Linder et al., 2017). However, these have come largely at the expense of user-friendliness. Aside from their inability to record how long a resource is used for, their excessively complex formulas and demands for data found in parts of the value chain outside the firm, diminish their use in practice. Finally, our metric is driven by ‘physical’ data, i.e. the number of times a resource is used. This is in contrast to financial or monetary data that many eco-efficiency measures use. The use of the latter is arguably insightful, but not as part of a circularity formula, as it distracts from the measurement of circularity per se. 5.4. Limitations and Future Research One of the main limitations of our approach is that refurbishment and recycling themselves require resources. Both the longevity indicator from Franklin-Johnson et al. (2016) that we make use of in this article, and our own circularity measure, do not take into account resource use in this way. Our approach shares this limitation with other approaches that concentrate only on particular parts of a life cycle. As Pauliuk (2018) argues by using the example of material substitution in vehicle manufacturing, there can be trade-off between the goal of using materials in a more circular way and life cycle benefits. What enhances a circular economy indicator might have a negative effect along the life cycle. Future research could perhaps develop circularity and longevity measures - especially in combination - that are combined with life cycle sustainability indicators (Pauliuk, 2018). We used gold to illustrate how our measure would work in practice, and it is easy to grasp how the measure would work for other resources that also have a long lifespan, and similar extents of circularity. Other 304 Ecological Economics 150 (2018) 297–306 F. Figge et al. resources, however, are markedly different. Similarly, our example of mobile phones has specific parameters for the potential number of times reuse and recycling can be pursued. Thus, future work could explore how our metric can relate to other resources with, for example, very different life spans. Further, different products could also be explored to develop the basis of our metric to accommodate multiple products - and different types of resources. Finally, because the scope of our indicators only extends to resources within specific product systems, other bases of circularity in particular cannot be accommodated. One example is open loop recycling, where resources may be used again but in different products. Whilst maintaining the objectivity of our indicator by only exploring processes and activities that the firm controls, our approach could be further developed into incorporating the recycling of resources that are transferred into other products or services that the firm generate. efficiency measurement for small and medium enterprise: a case study of wooden toy industry. Clean Techn. Environ. Policy 16, 935–945. Commoner, B., 1972. The environmental cost of economic growth. In: Ridker, R.G. (Ed.), Population, Resources and the Environment. Government Printing Office, Washington, DC, pp. 339–363. Daigo, I., Matsuno, Y., Ishihara, K.N., Adachi, Y., 2005. Development of Methodology for Analyzing the Average Number of Times of Use and the Average Residence Time of Iron Element in Society by Applying Markov Chain Model Tetsu-to-Hagané. 91. pp. 159–166. Daly, H., 1990. Toward some operational principles of sustainable development. Ecol. Econ. 1990, 1–6. Dubreuil, A., Young, S.B., Atherton, J., Gloria, T.P., 2010. Metals recycling maps and allocation procedures in life cycle assessment. Int. J. Life Cycle Assess. 15, 621–634. Eckelman, M.J., Daigo, I., 2008. Markov chain modeling of the global technological lifetime of copper. Ecol. Econ. 67, 265–273. Eckelman, M.J., Reck, B.K., Graedel, T., 2012. Exploring the global journey of nickel with Markov chain models. J. Ind. Ecol. 16, 334–342. Elkington, J., 1997. Cannibals with Forks: The Triple Bottom Line of the 21st Century Business. Capstone, Oxford. Ellen MacArthur Foundation, 2012. In Depth - Mobile Phones. Ellen MacArthur Foundation, Granta Design, 2015. Circularity indicators: an approach to measure circularity. In: Methodology. Ellen Mac Arthur Foundation and Granta Design. EU Commission, 2015. Closing the Loop-An EU Action Plan for the Circular Economy. Communication From the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions COM. 614. pp. 2015. Figge, F., Givry, P., Canning, L., Franklin-Johnson, E., Thorpe, A., 2017. Eco-efficiency of virgin resources: a measure at the interface between micro and macro levels. Ecol. Econ. 138, 12–21. Franklin-Johnson, E., Figge, F., Canning, L., 2016. Resource duration as a managerial indicator for circular economy performance. J. Clean. Prod. 133, 589–598. Geyer, R., Blass, V.D., 2010. The economics of cell phone reuse and recycling. Int. J. Adv. Manuf. Technol. 47, 515–525. Geyer, R., Kuczenski, B., Zink, T., Henderson, A., 2016. Common misconceptions about recycling. J. Ind. Ecol. 20, 1010–1017. Graedel, T.E., Allwood, J., Birat, J.-P., Buchert, M., Hagelüken, C., Reck, B.K., Sibley, S.F., Sonnemann, G., 2011. What do we know about metal recycling rates? J. Ind. Ecol. 15, 355–366. Grant, J., 1997. Planning and designing industrial landscapes for eco-efficiency. J. Clean. Prod. 5, 75–78. Green, D.I., 1894. Pain-cost and opportunity-cost. Q. J. Econ. 8, 218–229. Hagelüken, C., 2007. Metals Recovery From e-Scrap in a Global Environment, 6th Session of OEWG Basel Convention, Geneva. Hoffrén, J., Apajalahti, E.L., 2009. Emergent eco-efficiency paradigm in corporate environment management. Sustain. Dev. 17, 233–243. Huppes, G., Ishikawa, M., 2005. A framework for quantified eco-efficiency analysis. J. Ind. Ecol. 9, 25–41. Huppes, G., Ishikawa, M., 2009. Eco-efficiency guiding micro-level actions towards sustainability: ten basic steps for analysis. Ecol. Econ. 68, 1687–1700. Korhonen, J., Honkasalo, A., Seppälä, J., 2018. Circular economy: the concept and its limitations. Ecol. Econ. 143, 37–46. Krausmann, F., Gingrich, S., Eisenmenger, N., Erb, K.-H., Haberl, H., Fischer-Kowalski, M., 2009. Growth in global materials use, GDP and population during the 20th century. Ecol. Econ. 68, 2696–2705. Lai, J., Harjati, A., McGinnis, L., Zhou, C., Guldberg, T., 2008. An economic and environmental framework for analyzing globally sourced auto parts packaging system. J. Clean. Prod. 16, 1632–1646. Leontief, W., 1991. The economy as a circular flow. Struct. Chang. Econ. Dyn. 2, 181–212. Lewandowski, M., 2016. Designing the business models for circular economy—towards the conceptual framework. Sustainability 8, 43. Linder, M., Sarasini, S., Loon, P., 2017. A metric for quantifying product-level circularity. J. Ind. Ecol. 21 (3), 545–558. Lozano, F.J., Lozano, R., 2017. Assessing the potential sustainability benefits of agricultural residues: biomass conversion to syngas for energy generation or to chemicals production. J. Clean. Prod. 172, 4162–4169. Malthus, T.R., 1798. An Essay on the Principle of Population. Maxime, D., Marcotte, M., Arcand, Y., 2006. Development of eco-efficiency indicators for the Canadian food and beverage industry. J. Clean. Prod. 14, 636–648. May, G., Stahl, B., Taisch, M., 2016. Energy management in manufacturing: toward ecofactories of the future–a focus group study. Appl. Energy 164, 628–638. McDonough, W., Braungart, M., 2010. Cradle to Cradle: Remaking the Way we Make Things. North Point Press. Meadows, D.H., Meadows, G., Randers, J., Behrens, W.W., 1972. The Limits to Growth. Universe Books, New York. Müller, K., Holmes, A., Deurer, M., Clothier, B.E., 2015. Eco-efficiency as a sustainability measure for kiwifruit production in New Zealand. J. Clean. Prod. 106, 333–342. Paiano, A., Lagioia, G., Cataldo, A., 2013. A critical analysis of the sustainability of mobile phone use. Resour. Conserv. Recycl. 73, 162–171. Parajuly, K., Wenzel, H., 2017. Potential for circular economy in household WEEE management. J. Clean. Prod. 151, 272–285. Pauliuk, S., 2018. Critical appraisal of the circular economy standard BS 8001: 2017 and a dashboard of quantitative system indicators for its implementation in organizations. Resour. Conserv. Recycl. 129, 81–92. Pauliuk, S., Wang, T., Müller, D.B., 2011. Moving toward the circular economy: the role of stocks in the Chinese steel cycle. Environ. Sci. Technol. 46, 148–154. 6. Conclusions As society increasingly seeks bases of sustainable living, we are becoming more aware of the key responsibilities that organisations have. A spotlight shines on company behaviour, and reveals the importance of encouraging firms to use their resources as efficiently as they can. But how can they do this? Metrics that capture current rates of eco-efficiency are significant, but the development of approaches and tools that can inform better decision making to improve eco-efficiency are arguably even more important. The circular economy has become an increasingly popular idea for improving resource eco-efficiency. At the same time, the division between metrics that either adopt a circularity approach, or which emphasise longevity, highlights the incompleteness of both. In this article we have tried to combine them. In doing so, we exploit the individual strengths of circularity and longevity, and address the limitations of each. We propose that our approach represents a new direction for measuring and guiding resource eco-efficiency, and we look forward to others joining us in further endeavours. References Allwood, J.M., 2014. Squaring the circular economy: the role of recycling within a hierarchy of material management strategies. In: Worrell, E., Reuter, M. (Eds.), Handbook of Recycling: State-of-the-Art for Practitioners, Analysts, and Scientists., Waltham, MA, pp. 445–477. Alves, J.L.S., de Medeiros, D.D., 2015. Eco-efficiency in micro-enterprises and small firms: a case study in the automotive services sector. J. Clean. Prod. 108, 595–602. Bailey, R., Bras, B., Allen, J.K., 2004. Applying ecological input-output flow analysis to material flows in industrial systems: part II: flow metrics. J. Ind. Ecol. 8, 69–91. Bailey, R., Bras, B., Allen, J.K., 2008. Measuring material cycling in industrial systems. Resour. Conserv. Recycl. 52, 643–652. 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, 444–453. Blomsma, F., Brennan, G., 2017. The emergence of circular economy: a new framing around prolonging resource productivity. J. Ind. Ecol. 21, 603–614. Bocken, N.M., Olivetti, E.A., Cullen, J.M., Potting, J., Lifset, R., 2017. Taking the circularity to the next level: a special issue on the circular economy. J. Ind. Ecol. 21 (3), 476–482. Boulding, K.E., 1973. The economics of the coming spaceship earth. In: Daly, H.E. (Ed.), Towards a Steady-State Economy. Freeman, San Francisco. Boulding, K.E., 1996. The economics of the coming spaceship earth. Radical political economy: explorations in alternative. Econ. Anal. 357. Braungart, M., McDonough, W., Bollinger, A., 2007. Cradle-to-cradle design: creating healthy emissions – a strategy for eco-effective product and system design. J. Clean. Prod. 15, 1337–1348. Caiado, R.G.G., de Freitas Dias, R., Mattos, L.V., Quelhas, O.L.G., Leal Filho, W., 2017. Towards sustainable development through the perspective of eco-efficiency - a systematic literature review. J. Clean. Prod. 165, 890–904. Carvalho, H., Govindan, K., Azevedo, S.G., Cruz-Machado, V., 2017. Modelling green and lean supply chains: an eco-efficiency perspective. Resour. Conserv. Recycl. 120, 75–87. Cayzer, S., Griffiths, P., Beghetto, V., 2017. Design of indicators for measuring product performance in the circular economy. Int. J. Sustain. Eng. 10, 289–298. Charmondusit, K., 2017. Enhancing eco-efficiency of the kitchen furniture by using ecodesign. Asia-Pac. J. of Sci. Technol. 18, 677–696. Charmondusit, K., Phatarachaisakul, S., Prasertpong, P., 2014. The quantitative eco- 305 Ecological Economics 150 (2018) 297–306 F. Figge et al. Van Caneghem, J., Block, C., Cramm, P., Mortier, R., Vandecasteele, C., 2010. Improving eco-efficiency in the steel industry: the ArcelorMittal Gent case. J. Clean. Prod. 18, 807–814. Wen, Z., Meng, X., 2015. Quantitative assessment of industrial symbiosis for the promotion of circular economy: a case study of the printed circuit boards industry in China's Suzhou New District. J. Clean. Prod. 90, 211–219. Wiedmann, T.O., Schandl, H., Lenzen, M., Moran, D., Suh, S., West, J., Kanemoto, K., 2015. The material footprint of nations. Proc. Natl. Acad. Sci. 112, 6271–6276. Wilhelm, W., Yankov, A., Magee, P., 2011. Mobile phone consumption behavior and the need for sustainability innovations. J. Strateg. Innov. Sustain. 7, 20–40. World Business Council for Sustainable Development, 1996. Eco-Efficiency and Cleaner Production: Charting the Course to Sustainability. WBCSD, Geneva. World Business Council for Sustainable Development, 2000. Eco-Efficiency. Creating More Value With Less Impact. WBCSD, Geneva. York, R., Rosa, E.A., Dietz, T., 2003. STIRPAT, IPAT and ImPACT: analytic tools for unpacking the driving forces of environmental impacts. Ecol. Econ. 46, 351–365. Yuan, Z., Bi, J., Moriguichi, Y., 2006. The circular economy: a new development strategy in China. J. Ind. Ecol. 10, 4–8. Zhang, B., Bi, J., Fan, Z., Yuan, Z., Ge, J., 2008. Eco-efficiency analysis of industrial system in China: a data envelopment analysis approach. Ecol. Econ. 68, 306–316. Zhang, J.J., Joglekar, N., Heineke, J., Verma, R., 2014. Eco-efficiency of service co-production: connecting eco-certifications and resource efficiency in US hotels. Cornell Hosp. Q. 55, 252–264. Pauliuk, S., Kondo, Y., Nakamura, S., Nakajima, K., 2017. Regional distribution and losses of end-of-life steel throughout multiple product life cycles—insights from the global multiregional MaTrace model. Resour. Conserv. Recycl. 116, 84–93. Pearce, D., Turner, R.K., 1990. Economics of Natural Resources and the Environment. Johns Hopkins Univ. Press, Baltimore. Peng, H., Zhang, J., Lu, L., Tang, G., Yan, B., Xiao, X., Han, Y., 2017. Eco-efficiency and its determinants at a tourism destination: a case study of Huangshan National Park, China. Tour. Manag. 60, 201–211. Pimentel, D., Pimentel, M., 2006. Global environmental resources versus world population growth. Ecol. Econ. 59, 195–198. Potting, J., Hekkert, M., Worrell, E., Hanemaaijer, A., 2017. Circular Economy: Measuring Innovation in the Product Chain. PBL Netherlands Environmental Assessment Agency. Rochat, D., Hagelüken, C., Keller, M., Widmer, R., 2007. Optimal Recycling for Printed Wiring Boards (PWBs) in India, Proceeding of the R'07 World Congress, Recovery of Materials and Energy for Resource Efficiency Davos, Switzerland. Smol, M., Kulczycka, J., Avdiushchenko, A., 2017. Circular economy indicators in relation to eco-innovation in European regions. Clean Techn. Environ. Policy 1–10. Strasburg, V.J., Jahno, V.D., 2017. Application of eco-efficiency in the assessment of raw materials consumed by university restaurants in Brazil: a case study. J. Clean. Prod. 161, 178–187. United Nations Environment Programme, 2012. Global Environmental Outlook 5: Environment for the Future We Want. UNEP, Nairobi. 306