This page is referenced by:

• 1 2014-02-25T03:56:32-08:00 How do we know e-waste? Electronic discards and the double social life of trade statistics. 19 revpar 2014-10-02T05:25:06-07:00 Another important genre of knowing e-waste is the use of trade statistics. In a forthcoming paper (Lepawsky, 2014) I describe some of the key difficulties in measuring the international trade and traffic of e-waste using such data so that I can then go on and map their patterns (see here and here). I'll summarize those difficulties, but hold in the back of your mind that adjective 'international'. It might have rolled by you without much of a thought, but, as we'll see in a moment, when we're thinking about the social life of methods it is a little piece of language that turns out to be highly significant. The difficulties for measuring the international trade and traffic of e-waste are many, but can be summarized as follows:
  • There is no universally agreed upon definition of e-waste.
  Is a toaster with a digital clock 'e-waste'? Is a Toyota Prius? According to [oops... modest witness alert!] Manfred Broy, a professor of informatics at Technical University, Munich, a contemporary "premium-class" car "probably contains close to 100 million lines of software code" running on 70-100 microprocessors (Charette, 2009). The gamut of objects and materials that could potentially be considered waste of an electronic kind is immense, growing, and there are no non-arbitrary ways to include or exclude their legions. This inherent indeterminacy of what is or is not to be categorized as an electronic object makes coming up with robust statistical categories for them as a distinct class of objects to be recorded and counted by statistical agencies probably impossible.
  • No existing international trade data directly distinguishes between new and used electronics.
  Whether a piece of equipment that might be more readily recognized as an electronic item (e.g., a CPU, a monitor, a cellphone, a printer) is crossing a border because it is a new manufacture, fresh from the factory floor or old model sold to be reused again by some one else somewhere else is not directly distinguished in international trade data. Estimates must be made based on price thresholds that are themselves derivatives of price data available from sources other than international trade data itself (see Duan et al 2014).
  • Illicit and other forms of unrecorded or misrecorded trade aren't captured in trade data.
  This problem, while obvious in a certain way, is actually staggering in its scale well beyond the specific issue of e-waste (see Nordstrom, 2007 and here). The illicit and unrecorded is also, for the most part, invisible to the state and its statistical agencies, yet much of yours and my daily lives are dependent on those invisibilities. These three difficulties ensure that the use of international trade data make a variety of things visible, but at the cost of much, much more being or becoming invisible. But there is more here to be said, about the 'international', about what is going on methodologically when state-centric data are used to track e-waste. One of the key entities that are generated by international trade data are states 'themselves'. Data I use in several of my papers are from state based statistical agencies and collected by an international body: the United Nations Commission on Trade and Development (COMTRADE). In that data the fundamental units are 'territories', one cannot disaggregate to a more fine grained detail in such data. Most often, these territories in the COMTRADE data are states, but there are some non-state territories that are also interesting for how they flag up the 'state' as an arbitrary category by which to group data. A non-exhaustive list of examples of these other territorial groupings include "United States Minor Outlying Islands", "Other Asia nes (not elsewhere specified), "North Africa, nes", and "Occ. Palestinian Terr." (as in Occupied Palestinian Territories). The inability to disaggregate such data into other scales, especially intra-territorial ones such as 'the urban' has major consequences for how we know e-waste via trade data. Much of the e-waste debate is couched in terms of so-called 'developed' countries dumping waste in 'developing' countries. As such the issue of e-waste is framed at the scale of international relations. A consequence of that framing and all the data collected at that scale is that we are blind to the growing rise of 'middle-class' consumers within 'developing' countries, particularly in urban centres, that can and do purchase new and used electronics domestically - and have been doing so for some time - before subsequently ridding themselves of them; we're also blind to the fact that such sites have major institutional users of electronics - banks, hospitals, government offices - that are important domestic sources of discarded electronics. So by framing the issue of e-waste in terms of international trade we partially format the phenomenon in such a way that is not necessarily the most relevant to it. For example, a recent multi-method study of e-waste in Ghana - site of the infamous Agbogbloshie 'dump' - found that more than half of electronic discards in the country are from domestic sources, not imports of e-waste (Amoyaw-Osei et al. 2011).  More generally, we can say that trade data, in its statistical grouping work, forces us to consider what statisticians refer to in technical language as the 'ecological fallacy'. This is the error of making inferences about individuals that belong to a statistical aggregate - like a state - and where the individuals in question happen to be states themselves that have been aggregated into groups like 'developed' and 'developing' countries and the like. The latter issue raises a second: the modifiable areal unit problem (or MAUP), which amounts to a form of geographical and statistical indeterminacy (there are others, see Kwan, 2012) in which one can arrive at different summary figures (such as totals, rates, or proportions) that change based on how the boundaries are drawn to define group membership. The problem is those boundary definitions are arbitrary and cannot help but be thus. While these two issues might seem confined to the arid realms of statistical arcana, versions of them play out directly in real laws and policies designed to institute trade restrictions on e-waste imports/exports between various territorial groupings such as 'developed' and 'developing' countries, Organization of Economic Co-Operation and Development (OECD) and non-OECD countries, or under the Basel Convention, Annex VII and non-Annex VII countries (see Lepawsky and McNabb, 2010; Lepawsky, 2012; Lepawsky, 2014).
• 1 2014-07-07T06:48:13-07:00 Visualizing Transboundary Shipments of E-waste 8 Interactive cartograms of the e-waste trade. media 2014-07-07T07:49:04-07:00 How is the international trade and traffic of electronic waste organized? These cartograms are a partial answer to that question. 2012 1996 These visualizations are an interactive depiction of work published as part of the Reassembling Rubbish project in academic journals. For readers who may be interested, the full articles are available free without subscription at the following links:
  Lepawsky, Josh. 2014. “The Changing Geography of Global Trade in Electronic Discards: Time to Rethink the E-Waste Problem.” The Geographical Journal, April, n/a–n/a. doi:10.1111/geoj.12077. Click here for paper.
  Lepawsky, Josh, and Chris McNabb. 2010. “Mapping International Flows of Electronic Waste.” The Canadian Geographer 54 (2): 177–95. Click here for paper.

  ---

  What do the Cartograms Show?

  The cartograms depict data available from the United Nations Statistics Division's "COMTRADE" database. The data are organized into territories and the flows of e-waste that connect them. Territories are colour coded to reflect their status under Annex VII of the Basel Convention (blue for Annex VII territories and yellow for non-Annex VII territories). The larger the circle representing a given territory (e.g., Canada) the more other territories report receiving e-waster shipments from it. Readers not already familiar with the significance of the division between Annex VII and non-Annex VII territories are invited to download the publications above, both of which provide discussion about this issue.

  The flow lines are also colour coded to reflect the Annex VII status of their source. In other words, flows shown in blue originate in Annex VII territories; flows shown in yellow originate in non-Annex VII territories. The thicker the flow lines, the larger the volume of the flow.

  The data come with some extremely important caveats and limits. These are discussed fully in the Methods section of Lepawsky (2014, see above for free download). In summary, the caveats and limits of the data include:

  • The data showing information only for "waste batteries and scrap of primary batteries and electrical accumulators" (technically referred to as Harmonised System 2002 code 854810).
  • The data are import transactions reported by a given territory. In other words, these are data showing what Territory "A" says it receives as an import from any other territories in a given year.
  • The data miss unreported trade (whether licit or illicit).
  • The data visualized here are best understood as proxy measurements of overall transboundary shipments of electronic waste.

  How do I use the Cartograms?

  Interacting with the cartograms is as easy as using any webpage:

  • Click on a given territory (e.g., Canada) and its trade network will be highlighted and a side-bar with additional information will pop-up.
  • Read flow lines in a clockwise direction from a given territory to the territory(ies) it reports receiving shipments of e-waste from.

  How were the Cartograms Produced?

  A full discussion of the methods used to build these visualizations is available in Lepawsky (2014, see link above for free download). To summarize the process:

  • Data were downloaded from COMTRADE and organized in a spreadsheet.
  • The organized data were imported into Gephi, a free open-source network analysis software.
  • The networks created in Gephi were exported using Sigmajs Exporter, a project of the Oxford Internet Institute.