Downloads or Discs: which is greener?

Charles Corbett

One of the strategies that is often touted as a way to reduce our society’s dependence on finite resources is “dematerialization”. Loosely speaking, that refers to no longer selling physical products, but focusing on selling the services or information that those physical products were intended to deliver. Reading books or magazines on tablets and e-readers is one example. Car sharing, as a substitute for individual car ownership, is another. Downloading games over the internet instead of going to the store to buy a box with a disc in it falls in the same categry. But, how much greener is the download relative to the box? The answer may surprise you.

 

In a recent paper in the Journal of Industrial Ecology, Kieren Mayers and co-authors compare the carbon footprint associated with downloading a PlayStation 3 game with that of a Blu-ray disc. What exactly is a “carbon footprint”? Usually, that term refers to the greenhouse gas emissions that occur during production, use and disposal of a product, i.e., over its entire lifecycle. “Carbon” here is used as shorthand for “carbon dioxide-equivalent” (or CO₂-eq), as carbon dioxide is the most common greenhouse gas. So, what are the sources of emissions that Mayers and his co-authors consider in their analysis? For the Blu-ray disc, they consider the emissions associated with the initial development of the game, of the materials used to manufacture the discs, the emissions from the production process itself, and those associated with the distribution network, getting the discs to the retailers and into the consumers’ home. Then there are the emissions caused by consumers using electricity to play the game, and finally those associated with disposal of the disc.

 

For games downloaded via internet, some stages are the same as for those distributed on discs: game development, and consumers playing the game. However, instead of physical manufacturing and distribution, games for download are available through an online store, and downloaded via a broadband internet connection, both of which also consume energy.

 

The paper provides all the details of the assumptions the authors made, and precisely how they performed their comparison. To be precise, they work with the average size of a game available for the PlayStation 3 in the UK in 2010, which was 8.80Gb. They build on other studies to estimate the carbon emissions for each of the life-cycle stages for discs and downloads, and allow for the fact that the energy consumption associated with using the internet is subject to considerable uncertainty.

 

Taking all that into account, they estimate the life-cycle carbon emissions for the Blu-ray disc to be 20.82 kg CO₂-eq. To put that into perspective, an EPA report notes that the average vehicle in the US causes 411 g CO₂-eq per mile traveled, so the life-cycle carbon footprint of the Blu-ray disc is comparable to driving 5 miles in the average American car.

 

You’re still wondering how that compares to downloads. Surprisingly, Mayers and his co-authors find that, even taking low-end estimates for the energy consumption of the internet, the carbon footprint of the same 8.80Gb game downloaded over the internet is slightly higher, at 21.89 kg CO₂-eq. Looking at their detailed breakdown (Table 2 in their paper), two driving factors become clear. First, almost the entire carbon footprint of the Blu-ray disc, 93.6%, occurs during the game-play stage, and hence is the same whether the game was bought on a disc or downloaded from the internet. Second, the “internet” does contain a number of energy-consuming steps, including the equipment in the consumer’s home (such as modems and routers), the broadband access network, the core data transmission network, and the data centers. Even if not all of those are included in the analysis, they do add up.

 

One might argue that all that internet-related equipment is already in place and likely switched on anyway, regardless of whether the consumer is downloading a game or not. This may be true, and Mayers and his co-authors discuss this issue in more depth. This comes down to a core challenge in the field of “life-cycle assessment” (of which this particular study is one example): how to allocate emissions of a single process to the many products that rely on that process. In this case, the question is how to allocate the fixed emissions of the consumer’s modem and router and the other equipment to the games, music, books, magazines, searches, and other activities that all flow through that equipment. (For anybody familiar with cost accounting, this should sound like a familiar challenge: how to allocate fixed overhead costs to the different products using the same equipment.) In this case, Mayers and his team are not arguing that, as a direct consequence of downloading an 8.80Gb game, an additional 21.89 kg CO₂-eq will be emitted, but they are arguing that, if one tries to allocate all emissions fairly to all products and services that cause them, downloading is not environmentally “free” either.

 

 

References:

K. Mayers, J. Koomey, R. Hall, M. Bauer, C. France and A. Webb, “The Carbon Footprint of Games Distribution”, Journal of Industrial Ecology, 2014, DOI: 10.1111/jiec.12181.

 

Environmental Protection Agency, “Greenhouse Gas Emissions from a Typical Passenger Vehicle”, report no. EPA-420-F-14-040, May 2014, available at http://www.epa.gov/otaq/climate/documents/420f14040.pdf, last accessed October 30, 2014.