**4. Discussion**

In this chapter, we have provided an overview of the current state of WEEE regulation and collection trends in the EU and analysed and contributed to the consumer behaviour literature in the context of *e*CLSC by building a taxonomy of EEE categories. In this section we offer a discussion of implications, limitations, and avenues for future research.

In Section 2, discussing the EU's household WEEE related regulations, which is considered the state-of-the-art [4], the main implication from our research is that it is not enough. The increase in household WEEE collection falls behind the increase in new EEE put to the market, resulting in an ever-growing stream of WEEE that ends up being dumped or processed in the informal sector. This is not just an issue

in the EU, but a global issue as is evidenced in **Table 1** [9]. Considering that Europe is the region most effective in household WEEE collection, it is a suitable object of study when researching regulation and identifying novel ways of growing the urban mining sector. Bad WEEE collection is not just a missed financial opportunity [13], dumped or ill processed WEEE forms serious environmental and health hazards, hitting developing countries especially hard [10]. Based on these findings, we consider it critical that improvements are made in engaging consumers with proper and responsible WEEE collection and the role of consumers in *e*CLSCs should be well understood.

Some limitations of our work are related to the use of systematic keyword search for the consumer behaviour literature on *e*CLSC. In particular, by restricting our search to the context of CLSC, it identified mostly studies about the postconsumption or disposal phase. However, the existence of a recent literature review that identifies studies relating to all phases of consumption [35], reduces the negative impacts of the limitation discussed hereby. This is also supported by the similarity of our findings. Moreover, our search is able to identify a significant number of articles that are not identified by [35], hence offering a complementary view, rather than a biased one. Finally, to reduce potential negative impacts, we employed snowball mapping to identify further studies from the reference lists of the original set of articles.

We conducted a classification of consumer behaviours for *e*CLSC based on the consideration of direct, i.e. pertaining to considerations about equipment that has reached its EoU phase, and indirect, i.e. emerging from considerations about fulfilling some need(s), influences of consumer behaviour on *e*CLSC (see **Tables 3** and **4**). These have highlighted different levels of abstraction with which behaviours may be conceptualised. Future research could consider the concreteness of their conceptualisations to balance between reliability and generalisability of their results. While less concrete conceptualisations may offer less confident real-life predictions on a specific-case basis, results may be applicable to more scenarios. Moreover, abstract conceptualisations of behaviours (e.g. disposal) can be broken down into more concrete ones (e.g. return for incentives, sell secondhand, sell as scrap etc.) based on the objectives of the specific studies. While there are studies employing such techniques, e.g. [33], they do not do so explicitly. We have not encountered studies that also measure the overarching abstract behaviour in addition to its concrete constituents. Therefore, our findings suggest an interesting avenue for future research could be to account for these conceptualisations, of varying levels of abstraction, in order to allow for comparison of the results. In other words, *how do results change between the concrete and the more abstract conceptualisations of behaviours?*

The role of consumer behaviour in *e*CLSC was found to be critical to their success, but poorly understood by the surveyed literature. Specifically, there is a lack of variety concerning theoretical and methodological frameworks resulting in a significant amount of knowledge generation but pertaining to a very specific portion of the many simultaneous factors and processes leading to consumer behaviour. As such, institutional and cross-cultural perspectives, among others, have been overlooked in exchange for repeated applications of frameworks built around the TPB. Moreover, some studies employing this framework tend to assume that findings about behavioural intentions are realistically representative of actual behaviour, which has been widely rejected in the context of sustainable and ethical consumer behaviour. Therefore, we call for careful interpretations of such results.

In relation to methodological and theoretical innovation, we propose that, given the already existing literature on CLSC from a game theoretic perspective,

#### *Urban Mining of e-Waste and the Role of Consumers DOI: http://dx.doi.org/10.5772/intechopen.100363*

experimental economics [59] may offer an interesting framework for the empirical testing of extant and future models. Additionally, in this proposed direction, future research could focus on modelling and optimising of consumer decisions, on top of all other nodes in the modelled CLSC network. Moreover, the development of further structural models based on institutional and cultural considerations offers a natural continuation to the set of TPB-based studies.

Our review identified a sector of studies aiming to characterise the waste streams of certain national contexts through household surveys. While these studies offer little empirical explanation for the reasons behind observed behaviours, they do identify significant differences between the concrete behaviours, EEE categories and geographical contexts. This highlights the need for future research to work towards the elucidation of the most significant reasons for behavioural differences to emerge across all the aforementioned domains. To this end, we offer a conceptual starting point as discussed below.

In **Section 3.3** we developed a taxonomy of EEE categories by considering how their characteristics may predispose consumer behaviour towards different directions. While there had been some attempts to consider how EEE categories' characteristics tend to predispose consumers in specific behavioural areas [54], to the best of our knowledge, ours is the first attempt at capturing a large pool of dimensions through which to differentiate among EEE relating to all phases of consumption. The main benefits associated with our methods are those of replicability, extendibility and systematicity. Specifically, by following a fully transparent method and reporting on every iteration and decision made leading up to the final taxonomy [34], see Supplementary Materials document (see **Appendix**), we allow for plasticity and advancement of the taxonomy based, for example, on case-specific applications, theoretical and empirical criticism and extension or modification due to increase in data availability.

The classification of the sample of EEE categories used in the construction of the taxonomy was necessarily carried out *ad hoc* in some instances, due to unavailability of product-specific studies and data leading to potential ambiguity or disagreement among users. Therefore, it should be highlighted that the goal is not to set said classification in stone, but rather to provide a conceptually sound starting point for research to build upon as knowledge is generated. Despite its limitations, however, our taxonomy highlights what we have determined to be the most relevant characteristics that predispose consumers to adopt certain behaviours. We expect this to aid the inclusion of dimensions other than size in future comparisons of behavioural outcomes between EEE categories.

The main implications of our taxonomy can be summarised into the following points: (1) There are many physical, functional and symbolic dimensions and characteristics on which EEE categories may differ, leading to differences in behaviour. (2) There is a lack of research explicitly trying to consider and understand these differences, other than superficial explorations of those related to size and price heterogeneity. (3) Our result calls for further research to use, explore, expand and modify our taxonomy such that its value to researchers can increase. (4) There are interesting interconnections between some of the dimensions, which can be understood as moderation effects, which offer interesting avenues for future research. For example, value type may be able to explain some of the tendency to stockpile, but this may apply more strongly to small EEE categories than the other sizes, i.e. a moderation effect from size on the relationship between value type and stockpiling behaviour. (5) The EU would benefit from complementing extant classifications of EEE categories with our taxonomy's perspective in order to explicitly consider how consumers relate to EEE categories of different characteristics.
