A geographical analysis of CE

I am currently writing a concept proposal for a project to be submitted under the Horizon Call GreenDeal 3.2.. One of the questions to be answered is how big is a territorial unit for the purpose of a regional economy. It reminds me of two conversations. The first one is more than ten years ago, with a flax grower in Northern France. He said the primary extraction of fibre from the flax plant should happen at tractor distance of the field. Over 20 km distance costs to much money, time and CO2 emissions to drive with a pile of straw. Five years ago I visited the paper village of Eerbeek (NL). Eerbeek is 20 km of the nearest river, and it is a long time ago that the forests of the Veluwe were suitable for paper production. The answer was that Eerbeek has within a circle of 200km the largest concentration wealthy well organised citizens in the world. Hence it is a good point to process waste paper collected with 40 mln people.

Distance matters. The logic is of a first year geography, applying the theories of Hotelling, Christaller or Marshall, and it is called friction of distance. However to come back on flax, flax is extracted in two steps. In the first one (after field retting as happens in the picture above) is scutching and is a mechanical process of separating the fibre from the wooden parts of the plant is done for an area 20km of the field. It is called primary extraction. The output is a fibre. The fibre represents a smaller share of the plant and has more value and is transportable to a unit where the secondary process occurs. That is a process of cleaning, carding and wet processing. This can be done in a range of 200 km from the field. Hence 10-20 primary extraction site feed into a secondary extraction site.

But besides distance, economies of scale or agglomeration effects come into play. I did therefore mention Marshall as the father of the agglomeration effect theory. For flax the economics of scale arise at e.g. 5000 Tons pro year (in input). That requires 5000 Ha of land in a radius of 20km around the primary factory. That enables around 20 units between Brittany and the Netherlands. However assuming that 5000 Tons input gives 1000T of long fibre input and the minimal scale for a spinning unit is also 5000 Tons, a basin of 100.000 Tons straw produces in 20 units 20.000 T of spinnable materials, good enough for four spinning units. These are fictive calculations, but for the sake of the argument a spinning unit of 5000 Tons/year is a specialised unit working at costs making it necessary to sell the yarn as a quality and speciality fibre (above 5 Euro/kg). However the price of bulk yarns is rather around 2 Euro/kg. For bulk production the economy of scale arise above 100.000 Tons/year. For the example a polyester spinning unit below 25.000 Ton/year can only survive in specialty fibres (e.g. Diolen in Emmen), bulk production occurs in sites with more than 250.000 Tons (e.g. Reliance in Patalganga).

I guess that the logic of friction of distance and agglomeration effects also applies for waste processing and for other bio-based materials. It does not apply for digital services, but we should not allow digital products/services as proxies for materials. Materials are bulky and enquire bulky transport and bulky processes. For biobased processes the primary process is to extract lactose from a plant mass, the secondary process is to make a PLA. For waste the case is more complex, as in terms of polymers waste is a pile of plant fibre, animal fibres, synthetic fibres, contaminated with zippers, buttons, dyes, coatings, and elastane. It is much less a homogenous mass than the harvest of a flax field, of a potato field or a forest. Lets expand on the waste mass in relation to economies of scale and agglomeration effects.

To start of with a waste mass has two main currents, the upper current and the under current (onderstromen). The upper current is all the textiles that are resellable or re-usable as they are, without further processing. This is a tiny part of the waste mass: around 10% of the whole: branded clothing, high quality items, unstained and of a regular size. My trousers and jackets do not count, they have too short sleeves/legs for 97% of men (even if they come from quality brands). There is also a part that is good enough for Africa, roughly the undamaged jeans and sportswear. 90% is collected textiles do not have a value as such and need to be mechanically or chemically recycled. Mechanical recycling means breaking down to a fibre that can be spun. Chemical recycled means processing it to a polymer making it is extrudable. Even the streams that can be mechanically or chemically recycled is the minority of waste, since the majority has blends that are too costly to process.

For the upper current a liquid market is required; that means enough offer and supply making a successful match likely. To keep that in mind the upper current, the volume of high end clothing should be substantial enough, and their should be a demand. My guess is that only larger cities with a rich white collar class and a large student population have the critical mass for a good second hand market. In the Netherlands only Amsterdam qualifies, other cities lack depth and choice (to be verified). For the undercurrent value does not matter, but there is possibly economies of scale at play both for primary and secundary processing. For the secondary processing the size of a polyester or viscose plant will help, and since waste is low grade, the processing unit should be rather be at bulk size. So then, how many people supply enough waste to fed a polyester recycling unit or regenerated viscose (from cotton) unit of 100.000 Tons? It is likely to be around 20 mln people assuming that we have 5 kilo waste p.p.p.y. That is the entire Benelux, or the whole of Scandinavia. To feed a 100.000 tons unit, a number of primary sorting units is required. I have no idea of their scale (that needs to be researched), but if for sake of the argument 5000 Tons per stream/year is needed, primary processing needs regions of 1 million people preferably with a radius of 20-50km.

The calculations here above are all educated guesses. I look forward for all your comments, and possibly expressions of interest to be involved in a Greendeal 3.2. textile project.