Beyond PET: An extended Deposit-
Return System for plastic
packaging in Sweden
A qualitative investigation of challenges and
lessons from future and earlier Deposit-Return
Systems
MARCO SUTER
KTH ROYAL INSTITUTE OF TECHNOLOGY
SCHOOL OF ARCHITECTURE AND THE BUILT ENVIRONMENT
Beyond PET: an extended
Deposit-Return System for
plastic packaging in Sweden
A qualitative investigation of
challenges and lessons from
future and earlier Deposit-
Return Systems
MARCO SUTER
Supervisor
NILS JOHANSSON
Examiner
MIGUEL MENDONCA REIS BRANDÃO
Supervisor at Anthesis Group
LINUS HASSELSTRÖM
Degree Project in Sustainable Technology KTH Royal Institute of Technology
School of Architecture and Built Environment
Department of Sustainable Development, Environmental Science and Engineering
SE-100 44 Stockholm, Sweden
Imprint
Keywords
Deposit-Return System (DRS); plastic packaging;
Citation:
Suter (2019) Beyond PET: an extended Deposit-Return System for plastic packaging in Sweden. Master Thesis KTH, ABS, SEED, (Stockholm) Marco Suter
Degree Project in Sustainable Technology marco.suter@protonmail.ch
June 2019
KTH Royal Institute of Technology
School of Architecture and Built Environment
Department of Sustainable Development, Environmental Science and Engineering
SE-100 44 Stockholm, Sweden
than for other packaging materials in Sweden. In the context of a circular economy, higher plastic packaging recycling rates could contribute to increased resource efficiency and lowered greenhouse gas emissions. A deposit-return system (DRS) for plastic packaging could be an appropriate economic policy instrument to increase recycling rates. This thesis investigates if and how the scope of the DRS in Sweden could be extended to post-consumer plastic packaging. Potential challenges for an extended DRS are identified through semi-structured qualitative research interviews and a literature review. Solutions to overcome the challenges are identified with lessons from earlier DRSs and are complemented with other possible solutions.
The results show that the DRS scope could potentially be extended to small hollow bodies. This product group could be implemented gradually in existing infrastructure, is easy to grasp for consumers and its share of the plastic packaging waste is sufficiently high. The focus should be on non-food packaging to avoid hygienic issues. Higher consumer awareness concerning plastic issues and policies, which incentivize plastic packaging recycling, were identified as important drivers for an extended DRS.
Sverige än för andra förpackningsmaterial. Genom att öka återvinningen av plastförpackningar kan högre resurseffektivitet och lägre utsläpp av växthusgaser uppnås. Ett pant-system för plastförpackningar kan således vara ett lämpligt styrmedel för att öka mängden återvunnet material. Detta examensarbete undersöker möjligheterna för ett utökat pantsystem i Sverige där plastförpackningar är inkluderade. Potentiella utmaningar för ett utökat system identifierades genom semi-strukturerade kvalitativa forskningsintervjuer och tillhörande litteraturstudie. Inspiration och lärdomar av liknande fall användes för att minska effekten av dessa i synergi med andra alternativa lösningar.
Resultaten visar att användningen av ett pantsystem kan potentiellt utökas för att inkludera små ihåliga produkter. Denna produktgrupp kan implementeras med relativt små medel, inkorporeras i den befintliga infrastrukturen, förstås av konsumenter och står för en tillräckligt hög andel av avfallet för en mätbar effekt.
Fokus bör vara på icke-livsmedelsförpackningar för att undvika risk för kontaminering. Ökad medvetenhet hos konsumenter rörande plastfrågor i kombination med tydligare riktlinjer och styrmedel för att motivera återvinning av plastförpackningar identifierades som viktiga drivkrafter för ett utökat pantsystem.
Technology at the Royal Institute of Technology (KTH) in Stockholm.
First, I would like to thank my supervisors, Nils Johansson at KTH and Linus Hasselström at Anthesis Group, for the professional guidance and support as well as the Anthesis office in Stockholm for letting me write my thesis at their office and encouraging me to work on my Swedish small talk skills.
Second, I would like to thank the people that participated in this thesis as interviewees. This thesis would not have been possible without their interesting knowledge, reflections and thoughts from the recycling sector.
Lastly, I would like to thank Lucas and Simon for the proofreading and my girlfriend, family, and friends for being supportive throughout the whole thesis.
Marco Suter
Stockholm, June 2019
DRS Deposit-Return System
EPR Extended Producer Responsibility PET Polyethylene Terephthalate
RVM Reverse Vending Machine
WFD Waste Framework Directive (2008/98/EC) MCDA Multi-Criteria Decision Analysis
CBA Cost-Benefit Analysis LCA Life Cycle Assessment
GHG Greenhouse Gas
CO2-eq CO2-equivalents
MFA Material-Flow Analysis
1.1 Aim ... 3
1.2 Objectives ... 3
1.3 Disposition ... 3
1.4 Delimitations ... 3
2 Background ... 5
2.1 Theoretical background ... 5
2.1.1 Definition Deposit-Return System ... 5
2.1.2 Drivers for a DRS ... 5
2.1.3 Dimensions of a DRS ... 7
2.2 Impacts ... 14
2.2.1 Environmental impact of DRS ... 14
2.2.2 Economic impact of a DRS ... 16
2.3 Current conditions in Sweden ... 18
2.3.1 Plastic packaging policy in the EU ... 18
2.3.2 Plastic packaging in Sweden ... 19
3 Methodology... 25
3.1 Semi-structured qualitative research interviews ... 25
3.1.1 The interview procedure ... 25
3.1.2 Selection of Interviewees ... 26
3.2 Literature review... 26
3.3 Analysis of empirical material ... 27
3.4 Methodology discussion ... 28
4 Challenges to make a DRS extension feasible ... 29
4.1 Economic challenges ... 29
4.2 Behavioral challenges ... 32
4.3 Technological challenges ... 35
4.4 Organizational challenges ... 37
4.5 Material challenges ... 40
5 Possible solutions to identified challenges ... 41
5.1 Solutions to economic challenges ... 41
5.4 Solutions to organizational challenges ... 49
5.5 Solutions to material challenges ... 52
6 Conclusion ... 54
6.1 Roadmap for an extended DRS ... 55
6.2 List of recommendations to actors ... 56
6.3 Further research ... 57
References ... 59
List of Figures and Tables ... 66 Appendix 1: Interview guide ... I Appendix 2: Interview requests ...III
1 Introduction
After its introduction in the early 20th century, plastic consumption experienced huge growth rates and has rapidly entered every aspect of our lives thanks to its versatility and durability (OECD, 2018). But in recent years the general public became aware of the downsides of our plastic consumption. Especially the increasing use of single-use plastic has been heavily criticized due to the issues that arise when plastic waste ends up in ecosystems (OECD, 2018). Moreover, global plastic production consumes approximately eight percent of the annual global oil production. Four percent is used for plastic itself and four percent is converted into energy that is needed for plastic production (Fråne et al., 2012).
In the EU, 38 % of all plastic is used for plastic packaging (Worrell and Reuter, 2014). Due to the short life cycle of packaging, plastic packaging is responsible for 76 % of all plastic waste in the EU (Worrell and Reuter, 2014). In 2016, the emissions from production and end-of-life treatment of plastic packaging were responsible for just 1 % of EU's total CO2 emissions. With business-as-usual - increased demand and incineration - this share could increase up to 30 % of the EU's 2050 CO2 targets (Material Economics, 2018).
The Swedish plastic packaging collection system, which is a result of the extended producer responsibility (EPR), achieved a collection rate of 44 % in 2017 (SCB, 2018). Hence, at least 56 % was incinerated (SCB, 2018). The EPR scope for plastic packaging encompasses “packaging waste made from plastic that is not for beverages” (Riksdagsförvaltningen, 2018). The collection rate exceeded the national target of 30 %, which is defined by legislation (SCB, 2017).
However, the actual rate of material recycling is probably lower than 44 % due to rejects in the sorting process and losses in the recycling processes (Anderberg and Thisted, 2015; Svensk Dagligvaruhandel, 2018a). The rejects cannot be used for material recycling and are therefore incinerated.
Van Eygen et al. (2018a) performed a material flow analysis of plastic packaging in Austria in 2013. They concluded that the recycling rate is significantly lower than the collection rate due to low quality of the collected goods. Hence, the actual recycling rate in Sweden as well as in Austria cannot be stated with certainty.
In 2020 the government-mandated plastic packaging collection rate is increased to 50 % in Sweden (SCB, 2017). It can be expected that the goal for plastic packaging collection will be further increased in the coming decades. Therefore, an increase in the collection is necessary. As mentioned before, the actual recycling rate is considerably lower than the collection rate. Too much plastic packaging is incinerated instead of recycled, especially with regard to the EU CO2
targets for 2050 and the vision of plastic packaging in a circular economy (European Commission, 2018a; 2018b).
Increasing plastic recycling rates to create a circular economy can be desirable to increase resource efficiency, improve economic competitiveness, and lower resource dependency (European Commission, 2015). From a life-cycle perspective, however, recycling of plastic is not in every case the most desirable treatment option to achieve lower environmental impacts (Lazarevic et al., 2010). Lazarevic et al. (2010) suggest that greater environmental benefits from recycling can be achieved when plastic waste of high quality is recycled, which has low levels of organic residues and can substitute virgin plastic with a high ratio. Plastic recycling might also not be the most cost-effective treatment option in every case (Gradus et al., 2017).
It can be concluded that the collection rates, as well as the purity of the collected plastic packaging, has to increase in the following years in order to increase recycling rates. Reasons for the low collection and even lower recycling rates arise on several points throughout the value chain. In households, too much plastic packaging is not collected in the EPR system but rather as normal household waste which is then incinerated (Fråne et al., 2012). If the plastic packaging is collected in the EPR system, it might still be incinerated. This is a result of issues with sorting and the low economic value of recycled plastic (Svensk Dagligvaruhandel, 2018b).
To achieve both higher collection rates and better quality of collected plastic packaging a deposit-return system (DRS) for plastic packaging could be implemented. In a DRS, consumers pay a deposit upon purchased goods which will be refunded if the empty packaging of the good is returned. The high return rates of aluminium cans and PET bottles in the Scandinavian countries show that a DRS can result in high collection rates (Jordbruksverket, 2015). The DRS for PET bottles and aluminium cans was introduced in the Scandinavian countries to prevent littering (Jørgensen, 2011). Nowadays, in the context of plastics in a circular economy and resource efficiency, a DRS could be a valuable policy to achieve higher recycling rates. The Swedish DRS with high collections rates and a good public acceptance provides interesting circumstances to investigate, if the scope of the DRS could be extended to include plastic packaging.
There is a large body of existing literature about DRS for beverage containers and its benefits and drawbacks, especially from an economic perspective. Several countries discuss implementing a DRS for single-use beverage containers or implemented one in recent years (CM Consulting Inc. and Reloop Platform, 2016). For wider DRS scopes, however, there is little information to be found.
1.1 Aim
The aim of this thesis is to investigate if and how the scope of the deposit-return system in Sweden could be extended to plastic packaging and further identify challenges for the implementation.
1.2 Objectives
• Identify key challenges that have to be overcome to make an extension of the deposit-return system to plastic packaging feasible.
• Assess how these identified challenges could be addressed, with inspiration from the existing deposit-return system.
1.3 Disposition
This chapter gave a problem description, the aim, objectives, and delimitations of the thesis. The background for the investigated field of study in this thesis is given in chapter 2. Chapter 2.1 gives the theoretical background of DRSs in general, chapter 2.2 the environmental and economic impacts of DRSs, and chapter 2.3 an overview of current conditions in Sweden concerning plastic packaging and DRS. In chapter 3 the methodology, which was used in this thesis, is introduced. Chapter 4 and 5 present and discuss results from the empirical material regarding the two objectives. In chapter 6 the conclusions are shown, including a roadmap and a list of recommendations for different actors.
1.4 Delimitations
As mentioned in the introduction, there is only sparse research literature about DRSs for post-consumer packaging waste with a scope that goes beyond beverage containers. Consequently, the aim of this thesis is to explore if and how a DRS extension is feasible. As a result of this, an exploratory research approach is adopted for this thesis. A qualitative method, semi-structured interviews combined with a literature review, is in this case the preferable strategy since it allows the formulation of new hypotheses and a “relatively unstructured approach to the research process” (Bryman, 2012: p. 41) which would not be possible with a quantitative method.
This thesis investigates the DRS within the geographical boundaries of Sweden.
Sweden has a long tradition of DRS with high collection rates and acceptance in the public. These aspects provide good circumstances to investigate a scope extension of the DRS. Plastic packaging waste is investigated in a household context and not in an industrial context, where the quantity and composition of plastic packaging differ considerably (Stenmarck et al., 2018) . Plastic packaging waste is defined according to the Swedish EPR regulation: “packaging waste made from plastic that is not for beverages” (regulation 2018:1462 (Riksdagsförvaltningen, 2018)).
When discussing recycling in this thesis, it is generally referred to as mechanical recycling. With mechanical recycling, the plastic waste is separated, washed, ground and then pelletized (Ragaert et al. 2017). This differs fundamentally from the process of chemical recycling where plastic waste is chemically separated into its fundamental structures (Shen and Worrell, 2014). Regarding the feasibility of chemical recycling, Shen and Worrell (2014: p. 188) note that it “is technically possible, but the economic feasibility of large-scale operation is still to be proven”.
2 Background
The previous chapter showed that recycling rates of plastic packaging are too low to achieve a circular economy. Too much plastic packaging is incinerated instead of recycled. In this chapter, the general literature about DRSs, the impacts of DRSs, and the current DRS implementation in Sweden are discussed.
2.1 Theoretical background
In this chapter, the drivers for a deposit-return system (DRS) and its dimensions are introduced.
2.1.1 Definition Deposit-Return System
A DRS is an economic policy instrument, that combines a tax for consumption with a subsidy for returning the used good after consumption (Hennlock et al., 2015). The combination of an upstream tax with a downstream subsidy creates an economic incentive for consumers to return empty goods (Deprez, 2016).
Hence, higher collection rates can be achieved. A DRS is a commonly used policy instrument for waste treatment and is often used for homogenous products (Hennlock et al., 2015). In literature, DRSs are often also referred to as deposit- refund system, deposit-refund program, deposit return scheme. Lindhqvist (2000) categorizes DRS as a part of the extended producer responsibility (EPR) principle.
2.1.2 Drivers for a DRS
Based on earlier experiences, there are several drivers that can lead to an implementation of a DRS (Jørgensen, 2011; Tojo, 2011). In most cases, these drivers do not have an isolated effect but rather influence and amplify each other.
The most important drivers are listed as follows.
Policy
Policy frameworks are an important driver for the implementation of a DRS.
Policy frameworks can also influence decision-making in businesses. When the first nationwide DRS for aluminium cans in Sweden was implemented, the mitigation of littering was an important long-term goal (Jørgensen, 2011).
However, in recent years sustainable development and circular economy became ubiquitous topics and are important long-term goals in many policy frameworks such as the EU action plan for the Circular Economy (European Commission, 2015).
There are two possible types of legal requirements which can be used to implement a DRS (McCloughan, 2017):
• There is a direct legal requirement in the legislation which forces a certain industry to implement a DRS. This system is in place in Denmark, Estonia, Germany, Lithuania, and Sweden.
• The alternative option is to impose an indirect legal requirement. An example of this is Norway, where the taxes on beverages are lowered with increasing return rates (Sutton, 2018). This mechanism provides an incentive for the beverage producers to achieve high return rates since higher taxes adversely affects their sales. If the collection rate is above a certain threshold (95 % in the Norwegian case) the tax is omitted (Sutton, 2018). The industry can then decide on the policy instrument to achieve this threshold. In the Norwegian case, a DRS was chosen.
Consumers
Consumers can be a strong driver to improve waste management systems (Mwanza and Mbohwa, 2017). The littering issue, that initiated the DRS efforts in Sweden, was an important public issue and pushed decision-makers to act (Jørgensen, 2011). Viscusi et al. (2011) note that individuals can increase personal contentment from performing a pro-environmental action – recycling in this case.
Industry
McCloughan (2017) states that a DRS can only be successful if the affected producers, suppliers, and retailers support it. The role of the industry as a driver can be perceived as contradictory. On one hand, Numata states (2009) that it is hard to gain the support of private companies for a DRS. Reasons for this are extra administrative efforts, investments in infrastructure and ultimately higher prices on products (Numata, 2009). Examples for this opposition can be found in Germany before the implementation of the container deposit legislation in 2003 (Cantner et al., 2010). Can manufacturer, beverage producers, and retailers proposed a voluntary collection in order to avoid a DRS in Sweden. However, the voluntary approach did not meet the required recycling rate (Jørgensen, 2011;
Tojo, 2011). On the other hand, producers of juice and syrup products can include their products since 2015 voluntarily in the DRS (Olofsson, 2017). According to Returpack Svenska AB (2018a), 2% of their deposit products in 2017 were part of the system on a voluntary basis. This shows that the industry can support a DRS when the infrastructure already exists and a DRS is viewed as a good way to allow pro-environmentally actions for consumers.
In 1981 an environmental tax on aluminium cans replaced a fee on disposable containers in Sweden (Jørgensen, 2011). The tax was lowered when higher
collection rates were achieved. This created an economic incentive for producers and retail stores to achieve higher collection rates. As a result, the industry implemented a DRS (Jørgensen, 2011). This type of legislation turned the industry from an opponent to a driver for a DRS. It allowed to find a solution that achieves the set goal but on the industry’s terms.
2.1.3 Dimensions of a DRS
As seen in literature, one can define five dimensions for a DRS: economy, behavior, technology, organization, and material (Jørgensen, 2011; Simon et al., 2016; McCloughan, 2017). Therefore, they are used here as a framework to give a theoretical background on the DRS dimensions. These dimensions are interconnected, hence a clear distinction for many aspects is difficult.
Environmental aspects are not relevant as a DRS dimension but are an important result of a DRS (Simon et al., 2016). Therefore, they are discussed in the section about the impacts of a DRS, chapter 2.2. On the contrary, economic aspects are included as a dimension but also as an impact. On one hand, economic aspects, such as the economic incentive of a deposit, are vital as a dimension of a DRS (Deprez, 2016). On the other hand, the economic impacts are important too, to guarantee cost-effective policies (Deprez, 2016).
Figure 1: The five dimensions of a DRS, which are used as a framework in this report. The drivers and the impacts of a DRS are discussed in the background.
Economy
As explained above, a DRS is basically a combination of a consumption tax and a subsidy on the correct disposal of a used good. From an economic perspective, this mechanism attempts to internalize an externality which is generated through the wrong disposal of a used good. In regard to plastic packaging, the externality is caused by littering or incineration. This can be translated into an external cost: for example, the greenhouse gas (GHG) emissions due to incineration or the decreased recreational value due to littering can be expressed
in monetary values. This external cost is one that society has to bear and not the individual who disposed the used good (Deprez, 2016). The internalization of external costs should lead to a better outcome for society (McCloughan, 2017).
Concerning plastic packaging, the optimal outcome for society would be less littering, less incineration and more recycling (DEFRA, 2019).
Aalbers and Volleberg (2008) concluded that with a DRS optimal incentives are implemented for improving collection rates and decreasing illegal dumping, incineration, and landfilling. Several other studies show that a DRS can create optimal fee structures to achieve policy targets such as less illegal dumping or more recycling (Atri and Schellberg, 1995; Fullerton and Kinnaman, 1995;
Palmer and Walls, 1997; Walls and Palmer, 2001).
There are several alternative economic policy instruments that attempt to correct this externality. One instrument is a consumption tax on a product - a Pigouvian tax - which increases the price of a good to a socially efficient level (Aalbers and Vollebergh, 2008). However, a consumption tax does not incentivize the right disposal of a used good and is therefore not applicable to increase recycling rates (Deprez, 2016). There is also the possibility of introducing prohibitive regulation on unwanted disposal methods but the monitoring and enforcement costs would be very high (Deprez, 2016).
Behavior
If households pay for their waste management with a flat rate tax, an individual has no financial incentive to put effort into separating and returning household waste to separated fractions. 80 % of all Swedish municipalities have a flat rate tax on household waste in place (McCloughan, 2017). The implementation of a DRS creates an incentive for consumers to return their goods and redeem the deposit. Low (2012) notes that a DRS exploits people’s loss aversion and people are therefore willing to spend some of their time to redeem the deposits, even if it is an insignificant share of their income. This loss aversion further minimizes the occurrence of alternative disposal options such as burning or illegal dumping (Walls, 2011).
Thörnelöf (2016) notes that low-income groups generally recycle less than wealthier individuals. Therefore, the introduction of a DRS affects the recycling rates of low-income groups to a greater extent than high-income groups.
However, recycling behavior is also influenced by other factors than financial incentives. Abbott et al. (2013) showed that social norms can have a strong effect on household recycling behavior. Therefore, in a non-DRS context, social norms might be used to take on monitoring and enforcement functions. This can be taken advantage of with collection schemes such as kerbside collection, which
make recycling efforts visible and signal pro-environmental attitude towards the social surroundings (Vining and Ebreo, 1990; Oskamp et al., 1991).
Viscusi et al. (2011, 2013) investigated how recycling laws such as a DRS for water bottles influence recycling behavior and how these changes correspond with social norms and private values. Their results show that internal private values toward recycling have a significant influence on recycling rates. Contrary to the research mentioned above, social norms seem to have an insignificant influence on recycling behavior (Viscusi et al., 2011, 2013). Therefore, bottle deposits and stringent laws are less effective on individuals which already have positive values regarding recycling and recycle diligently. Viscusi et al. (2013) further state that individuals who were only recycling a few of their bottles without a DRS turn into thorough recyclers with a DRS in place.
A DRS can facilitate recycling for consumers since only deposit packaging is accepted at the reverse vending machine (RVM). However, it is still crucial that consumers can intuitively understand which packaging is part of the scope (McCloughan, 2017).
According to Miliute-Plepiene et al. (2016), convenience is the most significant factor for high household recycling rates. As a waste management system matures and its convenience increases, the effect of moral norms become less important. It is further noted that the effect of economic incentives as an important influence on recycling behavior could not be proven “with sufficient certainty” (Miliute-Plepiene et al., 2016: p. 49). However, Hage and Söderholm (2008) showed that the economic incentive of weight-based fees on household waste led to increased plastic packaging collection rates.
Producers might be opposed to a DRS since it causes additional administrative and labelling efforts (Gandy et al., 2008). McCloughan (2017) notes that producers could theoretically avoid a DRS by changing the type of packaging or reformulating their drinks.
Technology
As illustrated by Jørgensen (2011), technological progress in the form of RVMs was a major pillar of the success of DRS in Scandinavian countries. RVMs are automated devices which verify the returned good as part of the DRS (Jørgensen, 2011). They crush and sort the returned good and pay out the refund, which in most cases is a voucher. Throughout the years, the RVMs progressed also technologically with laser scanners for barcode identification and more efficient sorting (Jørgensen, 2011). The most recent technological developments are express-RVMs. The current RVM models are able to process only one returned good at a time, while the express-RVMs are capable of receiving larger quantities of goods at once and process more in a shorter time (Pantamera, 2019).
Technology is an important aspect to increase efficiency of a DRS. Moreover, it is also an important aid to prevent fraudulent behavior (McCloughan, 2017). In the beginning, the bottle shape was the distinctive feature, which had to be distinguished either manually or by RVMs (Jørgensen, 2011). For almost 25 years, barcode readers are used now to verify that a deposit was paid on the returned good and thereby prevent fraud with for example imported beverage containers (Jørgensen, 2011).
The application of RVMs in a DRS prevents misthrows, which keeps the fractions pure. This aspect combined with the economic incentive of refunds makes a DRS an instrument which can achieve high collection rates with a high material purity. In order to utilize these benefits from the DRS, well-developed sorting and recycling technology is necessary. Sorting and recycling technology for aluminium and glass is already quite sophisticated (OECD, 2018). This can be attributed to the fact that these fractions have been collected and recycled for decades but also to the relatively low complexity within these two waste streams (OECD, 2018). Plastic waste, however, is a more complex waste fraction which does not have such a long collection history and is often of low economic value (OECD, 2018). The increased awareness in plastic led and will lead to advancements in sorting and recycling technology (Garcia and Robertson, 2017;
OECD, 2018).
Organization
When a DRS is implemented, an organizational entity is needed to operate the system (McCloughan, 2017). Moreover, a common convention between all actors, which secures a certain degree of standardization of the involved products, has to be put in place (Gandy et al., 2008). The organizations that manage a DRS either emerge out of purely economic reasons (natural systems) or because there is a societal objective that is translated into a regulation (artificial systems) (Lindhqvist, 2000). For example, up until the end of the 20th century, it was more cost-effective for breweries to have a local DRS on reusable glass bottles in place than using single-use bottles (Jørgensen, 2011). These local DRS set up by the breweries would only have a relatively small geographic scope and consumers could often only buy new drinks after the empty bottles were returned.
Lower prices on single-use glass bottles, larger breweries that went beyond the local or national scale, and the emergence of cheap single-use plastic bottles made this system obsolete by the end of the 20th century (Jørgensen, 2011). For a very small-scale DRS, the administrative tasks might be relatively few due to the small number of actors involved. For a national or international DRS this can become quite complex and requires agreements and contracts (Gandy et al., 2008).
Concerning the actual implementation of a DRS, there are several important aspects that need to be considered since these can influence how the actual DRS works in practice. In artificial DRSs, these aspects are in most cases regulated by a governmental legal entity or a company. The following list provides an overview of the most important organizational aspects in existing DRS (McCloughan, 2017):
• The management system: A DRS can be managed centrally where a system operator is responsible for paying and refunding the deposit.
Consumers can get their refund at any store which is part of the system (McCloughan, 2017). A system can also be managed decentralized and thereby avoid a system operator. As a result, expenses are lower but consumers cannot get their refund at any store (McCloughan, 2017). This system is in operation in some states in the USA - New York, Michigan, Massachusetts, Maine, Iowa (McCloughan, 2017).
• The scope of the deposit: Most of the DRSs that are in place have glass, aluminium and/or tinplate, and PET bottles or one or two of the materials within the scope of their DRS (CM Consulting Inc. and Reloop Platform, 2016). Since the consumer is an important actor in a DRS, it is necessary that the scope of a deposit is easy to understand for consumers in order to achieve good return rates (McCloughan, 2017).
• The Level of the deposit: In theory, the level of the deposit could be calculated with economic models (Deprez, 2016). In reality, however, many of the variables are hard to evaluate in monetary terms. Therefore, it is difficult to calculate the most optimal deposit precisely. The deposit should be high enough to incentivize high return rates but not so high that it provides an incentive for fraudulent behavior (McCloughan, 2017).
Thörnelöf (2016) investigated the effect of the deposit increase in Sweden on metal cans in 2010. The results show that the increase had a low positive effect on collection rates in Sweden.
• The sales channels: When setting up a DRS, it is important to clarify which sales channels are included in the scope of the DRS. In the context of beverage containers, possible variations are that only products sold through retailers bear deposit or that all sales channels (e.g. bars, restaurant, takeaway food stalls) must have a deposit on their sold goods.
A further issue regarding sales channels is if imported and exported goods are also part of the DRS.
• The collection options: According to McCloughan (2017), it is common in North American DRSs to have designated return depots. In the Scandinavian countries, the retail stores are most often the place where consumers can get their deposit refunded
Material
In many countries, DRSs are applied to beverage containers to decrease littering and increase recycling rates (McCloughan, 2017). But DRSs can also be applied to materials such as batteries, used cars, and motor oil (Fullerton and Wolverton, 2000). In theory, DRSs could be applied to every material which could be a hazard after its use-phase. In this report, the focus lies on plastic packaging which has a large variety concerning polymer material(s), form, size, additives, and use cases. The OECD (2015) notes that it is impossible to operate a DRS in a satisfactory way with products that do not have clear definitions regarding material content, form, size, use case. Hence, a clear characterization of the product group or the definition of a more detailed fraction of the whole waste stream should be prioritized.
In the case of plastic packaging, one way to characterize the plastic packaging stream is according to the used polymers. However, the same polymers are used in a variety of products (Van Eygen et al., 2018b). This would result in a complicated system since it is not very transparent and comprehensible for consumers which polymers are used in a given plastic packaging item. On the contrary, a product-centered approach might be easier to grasp for consumers.
Van Eygen et al. (2018a) assessed the plastic packaging flows with a material- flow analysis (MFA) in Austria in 2013. The following product groups were used to assess the material flows:
• PET bottles
• hollow bodies small (e.g. liquid soap containers, ketchup bottles)
• hollow bodies large (e.g. laundry detergent containers)
• films small (e.g. potato chips packaging, bottle labels)
• films large (e.g. grocery bags)
• large EPS (expanded polystyrene, e.g. seafood transport boxes, electronics)
• other products within the plastic packaging waste flow
This classification could provide a good basis to assess the suitability of different plastic packaging for an extended DRS. There is no detailed information about plastic packaging flows in Sweden. However, it can be assumed that there are only marginal differences in Swedish and Austrian plastic packaging composition. Therefore, the data from the Austrian case by Van Eygen et al.
(2018a) can be used as a proxy. The following aspects are often discussed in literature and should be considered when assessing how to extend the DRS scope (Hennlock et al., 2015; McCloughan, 2017; Van Eygen et al., 2018b):
The volume of the DRS fraction is important. The larger the collected fraction, the more plastic can be diverted from incineration to material recycling.
Additionally, larger fractions benefit from economies of scale, in collection but also in the further steps downstream. If a DRS includes materials with more hazardous effect on the environment than plastic such as batteries or motor oil then the volume of the waste stream could become less important since the avoidance of any polluting emission has a much higher priority than an economically profitable collection system.
Besides the volume, the diversity within a fraction should also be incorporated.
In general, the material diversity within a DRS fraction should be kept as low as possible (Hennlock et al., 2015). Hennlock (2015: p. 98) notes that the extension of the DRS scope is “likely dependent on the possibilities of standardizing and homogenizing products”. More heterogeneous fractions lead to higher administrative costs (Hennlock et al., 2015). Having as few polymers as possible in a waste fraction is one way to homogenize the waste stream. This results in a streamlined system and further sorting and recycling processes are facilitated.
Potential hygienic issues are another aspect that should be considered. In the existing Swedish DRS, this is not an issue since there are almost no food remains present in the returned goods (Returpack Svenska AB, 2017). Any extension of the DRS should pose no hygienic issues due to bacteria growth for the consumers and the retail stores (Gandy et al., 2008).
In earlier DRSs, the focus was on mitigating littering (Jørgensen, 2011).
Nowadays, the increased collection and recycling rates are seen as further benefits of a DRS (McCloughan, 2017). The shift in focus from mitigating littering to recycling has consequences on the further processing of the collected goods. If the DRS’s main priority is prevention of littering, the subsequent waste treatment is irrelevant as long as the waste does not end up in the environment.
However, if the DRS is implemented to increase recycling rates, it is important that the returned goods have the correct material properties to be recycled (Garcia and Robertson, 2017). Otherwise, the effort to operate a DRS is nullified.
LDPE, PP, and HDPE are polymers for which sorting and processing technology is available and demand for these recycled polymers is high enough already (FTI AB, 2018). In contrary to PET trays, colored PET bottles, PP films, and PVC.
Sorting and recycling technology for these polymers is developed enough but the market is currently too limited for profitable recycling (FTI AB, 2018). PS and EPS are possible to sort but the volume of PS is too low for economic recycling and the technology to recycle EPS is not yet available (FTI AB, 2018).
2.2 Impacts
In the following chapter, existing literature on the environmental and economic impacts of planned or already implemented DRS’s are presented, analyzed and discussed. As stated in the introduction, there is almost no literature on DRSs with a wider scope than single-use beverage containers. When analyzing new implementations in a system, it is important that the environmental and economic impacts are investigated. This should help to avoid a shift of burden and allows to see if the policy achieves its environmental and economic goals.
2.2.1 Environmental impact of DRS
As explained in chapter 2.1.2, DRSs were historically used to ensure a closed loop of reusable bottles and to avert littering. However, the characteristics of a DRS make it an ideal policy tool to collect fractions which can be easily used for material recycling afterwards. The environmental benefits are therefore not coming from the DRS directly but rather from the increased recycling rates which are made possible through high collection rates and the pure fractions. In the Swedish context, the environmental impacts from a DRS for plastic packaging and increased recycling rates have to be compared to the present main processing method, which is incineration (Fråne et al., 2012).
In a comparative Life Cycle Assessment (LCA), Sevigné et al. (2017) investigated the environmental impact of an EPR system and a discussed DRS on single-use beverage containers in Spain. The studied impact categories were: abiotic depletion, acidification, eutrophication, global warming, ozone layer depletion, human toxicity, and photochemical oxidation. The functional unit is defined as one “ton of PET bottles, cans and beverage cartons ready to enter a recycling process” (Sevigné et al., 2017). The results of the LCA show that compared to the EPR system, the DRS results in an environmental impact reduction of 30 - 40 % for the impact categories abiotic depletion, acidification, eutrophication, global warming, ozone layer depletion, human toxicity. For the impact category photochemical oxidation, the environmental impact reduction of the DRS is 63 %. The lower environmental impact is a result of the lower share of rejects and the higher quality of the collected fraction which results in higher recycling rates.
It has to be noted, that the model used in the study assumes that almost all of the rejects from the sorting facility are landfilled.
In Sweden, the amount of municipal solid waste (MSW) ending up on a landfill is marginal while most of it is incinerated (Avfall Sverige, 2018). Depending on the polymer and its application, incineration is in some cases better than landfilling regarding global warming potential and in most cases better with regards to total energy use (Bernardo et al., 2016). This aspect decreases the transferability of Sevigné et al. (2017) results to the Swedish case.
Simon et al. (2016) performed an LCA of different beverage packaging materials with a focus on post-consumer bottle collection. The functional is defined as packaging material that is needed for providing packaging for 1000 l of drinks.
Their results show, that the collection of aluminium cans and PET bottles have the second-lowest emissions of GHG in a DRS, after kerbside bag collection.
Aluminium cans have GHG emissions of 1.1 kg CO2-equivalents (CO2-eq) for the functional unit of 0.33 l cans and 0.9 kg CO2-eq for the larger 0.5 l cans.
Collection with kerbside bags results in 0.5 kg CO2-eq for both sizes of aluminium cans. The collection of PET bottles with a DRS results in GHG emissions of 1.5 kg CO2-eq for 0.5 l bottles, 1.1 kg CO2-eq for 1 l bottles and 0.7 kg CO2-eq for 2 l bottles. With a kerbside bag collection for PET bottles, GHG emissions of 1.1- 1.2 kg CO2-eq are caused. Over the whole life cycle, the GHG emissions of aluminium cans were about 7 times lower with recycling than with incineration.
For PET bottles, the life-cycle with recycling caused about 4 to 6 times lower GHG emissions than with incineration. For glass bottles, however, a DRS is the collection system with the highest GHG emissions: 19.2 kg CO2-eq for 0.33 l glass bottles and 21.7 kg CO2-eq for 0.5 l glass bottles.
Lazarevic et al. (2010) reviewed several LCAs which compared mechanical plastic recycling to MSW incineration. The studied LCAs show that mechanical recycling is a favorable alternative, especially if the treated plastic waste stream has little organic contamination and consists of single polymer fractions.
Rigamonti et al. (2014) concludes that incineration of PET and HDPE is the worst-performing option for all investigated impact categories (global warming, acidification, eutrophication, photo-chemical ozone formation, ecotoxicity and human toxicity). A system with mechanical sorting and recycling achieves the lowest environmental impact in the most impact categories compared to other scenarios which include source separation and lower collection efficiencies.
Arena et al. (2003) note that the production of 1 kg recycled PET requires 42- 55 MJ of gross energy, while 1 kg of virgin PET requires 77 MJ. For PE, 40-49 MJ are required for recycling 1 kg and 80 MJ for the same amount of virgin PE.
As pointed out by Dinkel et al. (2017), plastic packaging collection schemes should achieve high purity levels in order to replace as much virgin material as possible.
In general, plastic packaging can help to reduce food waste (denkstatt GmbH, 2017). The additional GHG emissions from plastic packaging are in most cases significantly lower than the emissions from food waste packed in more traditional packaging such as paper bags (denkstatt GmbH, 2017). The net benefits are highest in high-value food products. The increased impact for the plastic packaging of a 330 g steak is 5 g CO2-eq. As a result of the longer shelf life,
730 g CO2-eq can be avoided. However, in some cases, especially vegetables, the additional GHG emissions from the plastic packaging outweigh the avoided emissions from food waste (denkstatt GmbH, 2017).
2.2.2 Economic impact of a DRS
In this chapter, the existing literature about the economic impacts of already implemented or discussed DRSs is presented. For a more convenient comparison, all the costs are also listed in Euro (€), the exchange rates of April 23rd 2019 were used.
Implemented DRSs
There is little information of the economic impact of the Swedish DRS for PET bottles and aluminium cans. One estimate by Hedelin et al. (2003) assessed the costs at around SEK 300-400 million per year with no estimates for the benefits.
The report mentions the cost for handling a single aluminium can of SEK 0.22 and for a PET bottle of SEK 0.82. With the return rates from 2016 (Returpack Svenska AB, 2017) an annual cost of SEK 909 million can be estimated (€ 86.4 million, SEK 1 = € 0.095). However, it can be assumed that the actual cost is lower than this sum. The cost is not expected to rise linearly with more returned goods. Rather should the handling cost per unit of returned good decrease with scaling effects.
In Germany, the DRS for PET bottles, aluminium cans, and glass bottles was introduced in 2003. The direct monetary costs - expenditures for personnel and infrastructure - were estimated at € 640 million with additional € 340 million in up- and downstream costs (Thörner et al. 2007). The report pointed out the high costs per reduced unit of CO2 emissions, however, achieving a significant CO2
emissions reduction was not the main reason for implementing a DRS. Apart from the avoided CO2 emissions,no statements were made about the magnitude of benefits as a result of the DRS.
Lavee (2010) performed a cost-benefit analysis (CBA) of the Israeli DRS for beverage containers and concluded that the DRS benefits are about 35 % higher than the costs. However, the fact that landfilling is the preferred option for waste treatment in Israel makes it difficult to compare these results to the Swedish case.
Moreover, Deprez (2016) points out that the benefits might not be as high as stated, due to the fact that some minor costs are not accounted for in the CBA.
A CBA was also performed by Vigsø (2004) for the Danish DRS. The major share of the costs in the DRS is from direct social costs such as handling costs, collection, transport, sorting, processing for reuse. Compared to incineration, the DRS and the subsequent recycling causes additional costs of € 6.7 to € 8.1 million per year that are borne by the consumers. The author notes that the study was
done shortly after the implementation of the DRS and therefore a future decrease in costs can be expected as a result of the system becoming more efficient.
Planned DRSs
Regarding DRSs that are not yet implemented but are discussed or planned, there are three noteworthy studies on the economic impacts of a DRS. The most recent is the one by the UK Department for Environment, Food and Rural Affairs (DEFRA) (2019). Two possible options were assessed for the public consultation;
an all-in option and an on-the-go option (includes only containers up to 750 ml).
The all-in option causes estimated discounted costs of £ 7’211 million (€ 8’365 million, £ 1 = € 1.16) for the first ten years. The benefits are estimated at £ 9,400 million (€ 10’905 million,), resulting in a net present value of £ 2,189 million (€ 2’539 million). For the on-the-go DRS, the discounted costs are estimated at
£ 2,764 million (€ 3’206 million) over the first ten years with benefits of £ 3,012 million (€ 3’493 million), resulting in a net present value of £ 249 million (€ 288 million) (DEFRA, 2019).
The costs of the discussed DRS alongside an EPR system for beverage containers in Spain was assessed by Fullana i Palmer et al. (2017). The following types of materials should be included in the DRS: PET, HDPE, steel, aluminium, carton for drinks, and glass. The estimated costs for just the DRS were estimated at
€ 1’800 million per year while handling the materials in the EPR would cost only
€ 165 million. The broad range of materials is assumed to be handled manually for the most and not with RVMs. The manual handling of the returned goods is responsible for 80 % of the costs of the DRS.
Dráb and Slučiaková (2018) investigated the costs of a possible DRS for single- use beverage container in Slovakia, which would be inspired by the Scandinavian systems. The investment costs are estimated to be approximately € 80 million, with € 62 million for the RVMs. The annual operation costs amount to
€ 33.3 million, while the revenues from unclaimed deposits and raw materials amount to € 28.3 million. The deficit of € 5.1 million will be paid by the producers via the administrative fee but in the end, probably carried over to the consumer. Dráb and Slučiaková (2018) point out two important aspects that the implementation of this type of DRS brings with: first, the operational revenue is dependent on the unclaimed deposits and if more deposits are refunded, the administrative fee must be increased. Second, the DRS will lead to a decrease in quantity but also in the quality of raw material in the already existing EPR system which will result in higher costs for the EPR system.
As the presented literature show, there is no coherent result of the costs and benefits of a DRS. One reason for the different results in the literature are the differences in waste management systems and possible system transitions in
different countries. Moreover, certain studies include costs and benefits which are not included in other studies and, as a result, end up with different conclusions. In general, the magnitude of costs is in most cases approximately the same when the size of the respective country is taken into account. Many studies include only the direct costs and benefits since there is relatively good information available. Indirect costs and benefits, such as reduced littering, GHG emissions, and time spent separating, are harder to assess and, therefore, in most cases not included. In general, it must be kept in mind that most of the (modeled) impacts are on the basis of present economic conditions. Stricter regulation in the future to curb GHG emissions and facilitate circular economy efforts could lead to a drastic change in the economic conditions. This would also impact the costs and benefits of DRSs.
2.3 Current conditions in Sweden
In this chapter, the waste management system in Sweden is presented. First, on the EU level and then on the national level from a policy perspective and then how these policies are put into practice.
2.3.1 Plastic packaging policy in the EU
Sweden is a member of the European Union (EU) and, therefore, has to adopt EU legislation in the form of regulations and directive. Up until now, there is no legislation that directly addresses plastic packaging and plastic packaging waste.
Plastic packaging is addressed indirectly through municipal solid waste (Waste Framework Directive 2008/98/EC) and packaging waste (Directive 94/62/EC and Directive 2004/12/EC) (Milios et al. 2018). The Waste Framework Directive introduced the European waste hierarchy which sets the prioritization of waste treatment: prevention, minimization, reuse, recycling, energy recovery, disposal.
Directive 2004/12/EC states that member states must introduce a collection scheme to increase collection rates (Ragaert et al., 2017).
However, plastic packaging waste is becoming a more important waste stream in the European Union. As a result, it is more directly addressed. The Green Paper on a European Strategy on Plastic Waste in the Environment (European Commission, 2013) investigated possible challenges for public policy that are caused by plastic waste. In 2015 the EU Action Plan for a Circular Economy was published (European Commission, 2015). The Action Plan focuses on introducing measures to facilitate a more circular economy in the EU and plastic packaging recycling plays an important role in that. The “European Strategy for Plastic in a Circular Economy” (European Commission, 2018b) further focuses on the key role that plastic has on the road to a circular economy within the EU.
The directive on single-use plastics bans certain single-use plastic items such as cotton buds, cutlery, plates, straws and wants to reduce the consumption of food containers, beverage cups, and containers (European Commission, 2018c).
The Packaging and Packaging Waste Directive set a minimum recycling target of 22.5 % for plastic packaging that must be achieved by the end of 2008 (European Commission, 2004). The target is specifically defined as “counting exclusively material that is recycled back into plastics” (European Commission, 2004: p. 28).
The revised directive increases the recycling targets to 50 % by 2025 and 55 % by 2030. Moreover, a clearer definition of the measurement point of recycling was introduced: “should be at the point where packaging waste enters the recycling operation” (European Commission, 2018d). These political decisions show a strong political will at the EU level to address the issues originating from plastic.
Moreover, they can create pressure on Swedish policy makers and industry to take additional actions.
2.3.2 Plastic packaging in Sweden
As explained above, Sweden must implement the EU Waste Framework Directive (WFD), which is done in the Swedish Environmental Code (Anderberg and Thisted, 2015). In the Environmental Code, it is stated that the municipalities are responsible for collection and treatment of MSW (Fråne et al., 2014) but since 1994 producers and importers of packaging are responsible for collection and treatment of their packaging waste.
Policy on EPR for Packaging
In 1994 the extended producer responsibility (EPR) for packaging was introduced on the basis of regulation 1994:1235 (Lilienberg et al. 2006). The EPR shifts the responsibility for packaging end-of-life treatment to the producer.
Therefore, the producer should have an incentive to include aspects of better recyclability and resource efficiency into their product design (Walls, 2006).
With an EPR in place, the producers are forced to introduce a collection system for their packaging waste and have to achieve set collection or recycling goals – depending on the legislative formulation (Riksdagsförvaltningen, 1994). As mentioned above, this regulation led to a shift in responsibilities for packaging waste from the municipalities to the producers. The scope of this EPR system encompasses paper, cardboard, plastic, glass, and metal (Riksdagsförvaltningen, 2018). However, packaging that is not sorted out by consumers but disposed in the normal MSW falls under the responsibility of municipalities and is incinerated (Hage and Söderholm, 2008).
Policy on DRS
As mentioned in chapter 2.1.2, the DRS for aluminium cans was not government- mandated (Jørgensen, 2011). The industry implemented the DRS as a means to achieve higher can collection rates. The incentive of a lowered environmental tax with higher collections rates was high enough for the industry to implement a DRS. Later on, the DRS became government-mandated (Riksdagsförvaltningen, 2005). All professionally bottled beverages need to be part of a government-
approved deposit-return system (Riksdagsförvaltningen, 2005). The definition of plastic bottles in 1994 included only beverage bottles made of polyethylene terephthalate (PET). Subsequent regulation changed this to bottles from polymeric materials (Riksdagsförvaltningen, 1994, 2005). Dairy products, vegetable, fruit or berry juices were exempt from this regulation due to hygienic concerns (Gandy et al., 2008). Since 2015 the Swedish DRS is open for vegetable, fruit and berry juices on an optional basis (regulation 2014:1073) (Axfood, 2015).
The collection target for the mandatory scope is set to 90 % (Riksdagsförvaltningen, 2006). Packaging that is in the DRS on a voluntary basis has to have a collection rate of 30 %, similar to the EPR goal (Riksdagsförvaltningen, 2018).
The EPR system in practice
In order to comply with aforementioned regulations, the private-owned business FTI AB was founded (McCloughan, 2017). It is affiliated with and owned by most EPR-affected producers (FTI AB, n.d.). TMR is another company that is managing a system to comply with EPR regulation (Konkurrensverket, 2019).
For plastic packaging in Sweden, the recycling goal is at 30 % of the sold packaging that have to be recovered or rather collected through the system (Riksdagsförvaltningen, 2018). Since the implementation of this regulation, the goal has never been revised. However, in 2020 this goal is increased to 50 % and is, as a result, more ambitious than the goals set by the EU (Anderberg and Thisted, 2015). In Figure 2, the amount of plastic packaging produced and collected per year, and the collection rate can be seen.
According to SCB (2018), 215’600 tons of plastic packaging was put on the Swedish market in 2017 and 95'500 tons were recovered through the collection
0%
20%
40%
60%
80%
100%
- 50'000 100'000 150'000 200'000 250'000
2012 2013 2014 2015 2016 2017
Plastic Packaging, produced [t] Plastic Packaging, collected [t]
Collection Plastic packaging
Figure 2: The collection rates of plastic packaging in the years 2012-2017 (SCB, 2018). The left axis shows the produced and collected amount of plastic packaging, the right axis shows the collection rate.
system that is administrated by the packaging producers. With a collection rate of 44 % the goal was achieved. However, to achieve a 50 % collection rate and probably even higher rates in subsequent years, some significant adjustments to the system are needed. Moreover, according to the industry organization of Swedish retailers (Svensk Dagligvaruhandel, 2018b) there is a significant gap between the reported rate and the actual rate of plastic that is recycled into new plastic products. It is estimated that about half of the collected plastic packaging is incinerated and used for energy recovery which results in an effective material recycling rate of around 25 % (Svensk Dagligvaruhandel, 2018b).
The industry organization of Swedish retailers published a road map to make all plastic packaging recyclable by 2022 and only use plastic packaging from renewable or recycled materials by 2030 (Svensk Dagligvaruhandel, 2018b). The roadmap includes an analysis of the current situation, investments in a new sorting plant, and introduction of fees based on the packaging’s recyclability to incentivize using more recyclable packaging materials. A DRS could be an integral tool to increase recycling which is needed to provide enough recycled material by 2030.
A major share of the post-consumer plastic packaging is not sorted and therefore incinerated as MSW. If the plastic packaging is sorted, the fractions of homogeneous waste are rather small and often contaminated with other types of plastics and separation can be too costly (Felix, 2015).
As mentioned before, there is no detailed data on the plastic packaging flows in Sweden. The information provided by Van Eygen et al. (2018a) is used instead to give an approximation of the plastic packaging flows. Figure 3 shows the percentage of the total plastic packaging waste stream for each product group. It can be seen that Films small and Films large have shares of over 20 % of the plastic packaging stream. PET bottles and Hollow bodies small have both have shares over 15 %.
Figure 3: Share of the plastic packaging product groups according to Van Eygen et al. (2018).
0%
5%
10%
15%
20%
25%
30% PET bottles
Hollow bodies small Hollow bodies large Films small
Films large Large EPS Other Products
The aforementioned Austrian MFA also contains information about which polymers are present in which product groups (see Table 1) (Van Eygen et al., 2018a). Apart from other products, all product groups have a relatively low polymer variability. However, the share of each polymer in a product group can differ drastically.
Table 1: The product groups and the polymers that are present in this product group according to Van Eygen et al. (2018). Note that the polymer LLDPE is a variation of LDPE.
Product group Polymers in product groups
PET bottles PET
Hollow bodies small HDPE, PP, PS Hollow bodies large HDPE, PP
Films small LDPE, LLDPE
Films large LDPE, LLDPE
Large EPS EPS
Other Products LDPE, HDPE, PP, PS, EPS, PET
The DRS in practice
In 1984 the aluminium can DRS was implemented and in 1994 the PET bottle DRS was approved by Jordbruksverket and introduced nation-wide (Pantamera, n.d.). Consumers in Sweden had been taught with information campaigns how to use the DRS when it was first introduced for aluminium cans (Jørgensen, 2011). Later then, when PET was added to the DRS, consumers were already used to it (Pantamera, n.d.). For the administrative tasks, such as finances, organizing transport, and information campaigns, the company Returpack Svenska AB with its two subsidiaries - Returpack-Burk Svenska AB and Returpack-Pet Svenska AB - was founded (McCloughan, 2017). The company ownership is shared with 50 % to Sveriges Bryggerier – the Swedish Brewery industry organization, 25 % to Svensk Dagligvaruhandel – the Association of large retail chains, and 25 % to Livsmedelshandlarna – the Association of small & individual retailers (Tojo, 2011). In theory, multiple different DRS could co-exist with independent collection systems (Jordbruksverket, 2015).
Returpack Svenska AB is not profit-oriented and has its own facility for sorting in Norrköping (Tojo, 2011). The expenditures are paid by unpaid deposit, interest that arises from the deposit, administrative fees, and the income from selling the raw materials (Pantamera, n.d.; Tojo, 2011).
Figure 4: A money/material flow chart of the Swedish PET Pantamera System. Adapted from CM Consulting Inc. and Reloop Platform (2016).
In Figure 4 the money and material flow in the Swedish DRS for PET (Returpack) can be seen, which is identical to the aluminium can DRS. The collected cans are mostly recycled into new aluminium products since aluminium has no quality loss when recycled (Haupt et al., 2017). The collected PET bottles, on the contrary, are in the best case recycled into new bottles (Haupt et al., 2017). If the quality of the returned bottles is inferior, they are used for other plastic products with lower material requirements or are incinerated (Haupt et al., 2017). Besides Sweden, Croatia, Denmark, Estonia, Finland, and Norway have exactly the same model implemented (CM Consulting Inc. and Reloop Platform, 2016).
In the Technical Specification and Marking Manual, the required specifications, such as dimensions, polymer, material thickness, and barcode marking, for the bottles are stated (Returpack Svenska AB, 2018b). Besides PET, HDPE and PP bottles are in theory permitted but according to Bergendorff (2019a), no other polymers than PET are in the DRS at the moment.
In 2017 approximately 4’100 different cans and plastic bottles were part of the Swedish DRS and another 80 products were part of the deposit-return system on a voluntary basis (Returpack Svenska AB, 2018a). Figure 5 shows that the mandatory products of the DRS never achieved the collection goal in the years 2012-2017. In 2017, 21’300 tons of PET bottles and 16’600 tons of aluminium cans were collected with the DRS.
As shown in Figure 4, retailers are the point of collection for the used bottles. In most cases, this is done at one of the 4’000 automated RVMs, while only a small percentage of retailers handles the deposit without RVMs (Returpack Svenska AB, 2018a). The retailers receive a handling fee for every returned bottle or can.
Moreover, the RVM is viewed by most retailers as a way to bring customers into their store which should lead to higher sales. (Pantamera, n.d.).
0%
20%
40%
60%
80%
100%
- 5'000 10'000 15'000 20'000 25'000 30'000
2012 2013 2014 2015 2016 2017
PET bottles, produced [tons] PET bottles, collected [tons]
Collection rate PET bottles
Figure 5: The collection rates of PET bottles in Sweden from 2012-2017 (SCB, 2018).
3 Methodology
In order to investigate how the DRS could be extended to other post-consumer plastic packaging in Sweden, several steps and methodologies are needed. As mentioned before, the explorative nature of the aim requires a qualitative research method to develop new insights and hypotheses within a topic for which only sparse research literature exists (Bryman, 2012). By conducting interviews with actors from the recycling industry, new knowledge about challenges for plastic packaging and lessons from the existing DRS is gathered. This knowledge is then compared and linked to existing knowledge from literature. This approach allows to investigate if and how an extended DRS can be implemented.
3.1 Semi-structured qualitative research interviews
The interviews have been conducted as semi-structured qualitative research interviews. The character of semi-structured interviews allows to compare answers from different interviewees but also to deepen a specific topic that the interviewee brings up, but was not thought of by the interviewer (O’Leary, 2014).
By interviewing key actors from different sectors, it is possible to collect information about different aspects and from different perspectives.
3.1.1 The interview procedure
An interview guide was prepared prior to the interview (see Appendix 1). An interview guide provides a common frame of questions and topics that should be discussed. Moreover, it ensures comparability between the interviews while at the same time leaving room for further probing by the interviewer (Patton, 2002). The interview guide was structured into the following parts: introductory questions, general questions about recycling, questions regarding the present DRS, questions regarding an extended DRS, and questions about lessons from the PET DRS implementation. To gain further insight into interesting aspects, follow-up questions and probing questions were asked. Before beginning with the interviews, the interviewees were informed about the purpose and structure of the interview. The participants had the opportunity to remain anonymous but all interviewees agreed to their name being published.
The interviews were all performed over the telephone, except for one which was done by e-mail (see further below). E-mail interviews were the least preferred option since they do not allow for further probing (Rowley, 2012). All the interviews were recorded in agreement with the interviewees. The audio recording allowed transcription of the interview and further analysis. The interviews took between 25 and 50 minutes, were done in English and were all performed between March 14th 2019 and May 8th of 2019. After the interview, the recording was listened to again and compared to the notes taken during the interview.
