Circular infrastructures: Substitution or addition?

Infrastructures are the invisible base on which all other structure is built. Like the roots of a tree their form does not predetermine how the tree looks like, which shape the leafs take and the color of the branches. Instead, there is a back and forth between the roots and the visible parts of the tree, as there is a mutual adaptation of all parts of the organism. However, knowing only abut the roots we can derive certain features of the tree, such as its size and probably more, if we are a biologist familiar with the movements of substances from the roots through the other parts of the tree. Similarly, for the electrical grid, an electrical engineer will be able to read the particular layout of transformer stations and the wires connecting them and then have a general idea about the buildings and their uses built upon this base. This counterintuitive exercise - to pretend the highly visible does not exist and has to be derived from something that is hidden from common view - is only possible for the expert who is trained specifically to treat infrastructures as something which is designed and manipulated directly: as tools and devices. This is part of what was called infrastructural inversion, foregrounding the background, a perspective that is at the core of every expertise in modern societies.

Being increasingly involved in research on the so-called circular economy, and given my long-standing interest in socio-technical infrastructures, I wonder from which infrastructural form we could derive the nexus of industrial production and consumption, and how this would look like in a circular form. Let’s start with the industrial paradigm which since its beginnings in 18th Century England has spread to span the globe. At its core is the industrial production site, a factory which sucks up human and material resources and transforms them in industrial production into consumer goods - using a lot of energy in the process. On the infrastructural level this requires access to stable energy as well as being a node in a tight web of transport connections. The latter is also important for the distribution of the products. Factories also produce waste, but they have an incentive to minimize the production of byproducts in the name of efficiency or to move them into other value chains. A similar process is at work on the side of the resource influx: In most cases, raw material is only one form of input, complemented or maybe even replaced by modules fabricated by other, smaller production sites that feed their products into the larger factories. This results in a cluster structure, where large nodes are surrounded by smaller ones. Indeed, the support of industrial clusters is a popular mode of industry policy. But back to the end-products. They are moved, often over large distances to the consumers, who, again need access to energy and transport ways, but now in much smaller dimensions, much more distributed. In consumption, then the product is transformed into waste, which is deposited in one way or another, now with no incentives to minimize byproducts. On this side of the cycle, we see transformation into energy, but also local landfill, some recycling and - as a particular perversion of global flows - the movement of waste sometimes over large distances until it is deposited and left to rot. So far the industrial, non-circular infrastructure. What about circularity?

On the grounds of what was said so far the biggest difference in a circular economy would be increased reuse on the side of the consumer, who lacks the incentive to capture efficiency potentials during the consumption process. Quite the opposite: Rapid product cycles and increased demand increase production volumes on the production side, which is highly desired in the the industrial paradigm. In terms of infrastructural form we can imagine two outcomes: First, less consumption of industrial products creates a weaker dependence on production clusters and decouples networks of consumption from their production counterparts. Second, recycling in the form of feeding consumption byproducts into new value chains may do the same: decoupling from industrial clusters, but only if new the production facilities are decentralised and moved closer to the consumer in the form of local re-use. Just as well waste as resource could travel back to production centres, which would add to the global movement of goods a return channel. In this scenario consumption becomes even closer coupled to production, although most likely via new intermediaries specialised in transforming waste into something that can act as input to industrial production. I propose to call this option “added circularity” as opposed to “substituting circularity” in which decoupling happens.

The decoupling option - if actually realised in infrastructural form - would weaken the centrality of industrial production considerably. Slower product cycles, local reuse, new production close to consumption based on consumption byproducts, all this would make for a very different design of transport and production infrastructures. However, given the missing incentives for consumers, less efficiency in smaller-scale, decentralised production and the cost of change, the second option is more likely. Here, additional recycling nodes and even more flows of goods (in the form of consumption byproducts) are added to the existing infrastructure with much less benefits for the environment. Or to return to the tree-metaphor from the beginning: In substituting circular infrastructures we could expect see a very different tree to grow, while the trees rooted in added circularity will see very much like their non-circular brothers.

Thomas Berker
Professor of Science and Technology Studies