The idea of a world based on active transport, and on cycling in particular, is a recurring theme in thinking on degrowth. This was one of the proposed transformative paths of the Manifesto of the Mouvement québécois pour une décroissance conviviale and this notion also plays an important role in the reflections of the Degrowth.info group, based in Germany. The mainstream media also associate degrowth with cycling.
Most degrowth advocates agree that the bicycle is a useful and desirable tool in a post-growth world, although some favour the promotion of walking. One of the precursors of the philosophy of degrowth, Ivan Illich, describes the bicycle as the ecological machine par excellence:
The bicycle and the motor vehicle were invented by the same generation, but they are symbols of two opposing uses of modern advancement. […] It is a wonderful tool that takes full advantage of metabolic energy to speed up locomotion. On flat ground, the cyclist goes three or four times faster than the pedestrian, using five times less calories.
French engineer Philippe Bihouix, for his part, sees it as an example of a low-tech machine, despite the relative technical complexity involved in its manufacture. Even a simple model, he points out, contains several hundred technically complex basic parts, which are difficult to produce locally. The processes include metallurgy of alloys and different metals, the machining and fitting of parts, vulcanizing tire rubber, producing anti-corrosion paints, and grease for the chain. Once built, however, “it is clearly possible for ordinary people to fully understand how it works, to tinker with it […] to keep it in good condition for many years, not to say almost indefinitely” (translation).
Some currents of degrowth stress the need for production at the local level, ideally through worker self-organization. These ideas can be found in the writing of Yves-Marie Abraham, for example, who argues that the production of the basic necessities of life should no longer be undertaken by private enterprise or the State, but by communities organized according to the principles of self-production and sharing of the means of production. As many other organizations, Polémos, an independent research group on degrowth based in Montreal, also advocates a form of work organization based on cooperative and commons-based production.
But what exactly is involved in the production and maintenance of a bicycle in terms of work organization, material and energy resources, as well as technical choices? This study looks at the form that bicycle production could take in a degrowth context and the dependencies that this mode of transportation entails. It considers the concept of the bicycle workshop, a facility that is smaller and more user-friendly than a modern factory, and interrogates the tensions between the simplicity of manufacture to be achieved and the technical efficiency necessary for a light manufacturing process.
Why efficiency matters
The issue of efficiency may seem too focused on industry and productivity to be a legitimate concern as far as degrowth is concerned. The drive for efficiency is sometimes associated with Fordism and the alienation of workers and society in general. This is not the intention here. The first concern must be to select techniques that are appropriate to the desired scale of production of bicycles in order to avoid wasting materials and energy. In a degrowth scenario, producing bicycles is also about providing a product that citizens can afford, especially in a context where people are choosing to use fewer resources.
The stakes are significant. When bicycles gained popularity in France, between 1890 and 1895, their purchase price represented 800 times the average hourly wage in that country. Given the current average wage of CAN$ 26.65 per hour in Québec in 2019, those 800 hours represent $21,320, about the same price as a small economy car. For this reason, the bicycle was first adopted by the bourgeoisie: doctors, notaries, country priests, as well as by public services, in particular the police and the army. It didn’t really become widely accessible until after the First World War. Due to the improvement in manufacturing techniques, its purchase cost represented only 200 hours of work in 1925 and 95 in 1957.
Energy specialist Vaclav Smil observes that modern techniques allow less energy to be used to achieve a given amount of work. The first steam engines, for example, transformed less than one percent of the embedded coal energy into useful mechanical work. In 1900, the efficiency of locomotives did not exceed 10% and that of the best compound steam engines, 20%. Nowadays, combined cycle steam turbines that transform gas into electricity have an efficiency of up to 60%. A modern hydraulic turbine can transform up to 95% of the kinetic energy of water into electrical energy.
The technological level therefore has a real impact on the amount of energy needed to accomplish a job. Conversely, a return to more archaic technologies would likely increase the energy consumption required to do the same job. As far as energy is concerned, the challenge, from a degrowth perspective, is to produce the necessary goods with as few energy resources as possible, while avoiding the trap of using the same amount of resources to make more – often superfluous – products.
The components of a bicycle
A modern bicycle can have up to a thousand parts, including screws and chain links. Several are simple and do not represent any particular industrial challenge. Here we will focus on a limited number of major components, emblematic of the manufacturing techniques to be used. We will then discuss the materials required before considering the simplest form of work organization suited to the manufacture of each component.
The first major component is the frame. Steel or aluminum is extruded to produce a hollow tube. A piece of metal is heated to soften it, then pushed through a mold using a powerful hydraulic press developing a force of hundreds or thousands of tons. It is also possible to shape flat metal into a roll and weld it at the junction, but this is a laborious and inefficient process. A few artisans are currently experimenting with bamboo and ash frames. This might prove to be a satisfactory solution holding promise for a degrowth scenario.
The production of ball bearings, used to reduce friction in the pedal axle and wheels, is an old technology. It seems that the first bicycle to use it was the Grand Bi in 1869. The casing is made from a steel tube that is cut into sections before being forged. The main challenge is making perfectly spherical balls out of very hard metal. The process starts with a heavy gauge steel wire that is roughly cut and formed before being heated and polished into balls between two cast iron plates. It is a specialized task that cannot be performed in a general machine shop, but can be done in modest sized workshops. A large global manufacturer of Swiss origin, Bossard, has 2,500 employees at 77 sites, that is 32 employees per plant.
Sprockets and chainrings, as well as the chain itself, are forged. This is essentially a matter of cutting the metal and giving it its final shape using machine tools. The metal is then tempered to harden it and to increase its wear resistance. The work is comparable for the derailleur mechanism, where it exists. These are relatively simple operations that can be carried out in a small machine shop equipped for the tasks.
Other metal parts, such as brake components and mudguards, are the easiest to manufacture, mostly using simple presses to shape the metal. The assembly of the wheels, and in particular of the many spokes, is normally entrusted to an automated machine. It may be necessary to revert to a more labor intensive hand assembly method.
Steel is the most important material used in bicycle manufacture. The frame has traditionally been made of a relatively flexible alloy based on chromium and molybdenum, while the parts subject to wear (ball bearings, chain, chainrings and sprockets) are made of very hard, high grade chromium steel and, depending on the specifications, on a small proportion of other metals such as vanadium and cobalt. It is possible to produce these alloys from recycled steel by adding the desired metals, provided that the starting steel is of compatible formulation (i.e., does not contain other metals that would compromise the required qualities of the alloy).
The manufacture of modern bikes generally uses aluminum for the frame. This metal is easier to work with and recycle than steel, but it is not considered to be as durable for a bicycle. It might not be the best choice for a low tech bike that is meant to be used for a very long time. On the other hand, it is a good choice for all kinds of minor parts, such as brake handles and mudguards.
We must rule out carbon fiber: it is complex to produce and very energy intensive, ranging from 51 to 79 kWh of energy per kilo. In comparison, the production of new steel requires 6 to 14 kWh and that of recycled steel, 2 to 4 kWh. The production of new aluminum from the ore requires 63 to 95 kWh, but only 3 to 5 to recycle it. We do not know how to recycle carbon fiber and its lifespan as waste is measured in centuries.
The other critical material is rubber, which is used in the manufacture of tires. It is mostly made of natural rubber from the rubber tree (hevea) native to South America. These days, production is concentrated in Southeast Asia. In a post-growth context, it would still be possible to import it, but difficult to transport it over long distances, and more equitable trade arrangements would be necessary. A local solution is possible in the form of the dandelion. It is easy to grow in monoculture, and its latex has been used as a source of rubber in the past, especially in the USSR during WWII. This resource is currently attracting interest and an industry could emerge in the coming years.
Thus, the manufacture of a basic bicycle that is both simple and durable requires relatively sophisticated materials, as well as the use of machine tools for the manufacture of various types of parts. These technical problems were solved at the end of the 19th century and it is possible to produce quality bicycles with simpler means than those currently in use. In particular, it is possible to do without digitally controlled robots and machine tools, whose future seems uncertain in a post-growth world. It is also conceivable to simplify the bikes themselves by sacrificing a little performance and by accepting a little more weight.
It is not clear whether the post-growth world will still rely on centralized electricity grids or whether it will be based instead on renewable sources organized into local grids. There will probably be less energy available than there is today and it will likely vary depending on the weather and the output of solar panels or wind turbines.
Production could therefore be limited to days when weather conditions are favorable. Intermittent production would require an organization to switch between working days and holidays depending on the amount of energy available. An alternative would be to locate the workshops near streams, where small hydraulic turbines could deliver constant, but less abundant, energy. This would have the effect of limiting the maximum size of workshops.
The production of new aluminum requires enormous amounts of electricity, which would presumably be of hydroelectric origin. The process can never be interrupted – the molten metal would warp the vessels as it solidifies – and intermittent energy sources are therefore out of the question. Aluminum can be recycled in arc furnaces as a batch process, but the electricity requirement remains high and cannot be met by small local installations. The same goes for recycled steel.
The production of new steel currently requires the use of coal or, more rarely, natural gas. These fuels not only provide the heat required for the process, but also carbon which combines with the oxygen in the iron ore in the form of CO2, leaving pig iron as result. There is also an alternative process based on hydrogen. It combines with the oxygen in iron ore to remove it as water vapor, which does not emit greenhouse gases. However, the production of the needed amount of hydrogen by electrolysis requires a lot of electricity. A post-growth world will therefore have to make sparing use of steel, recycled or new.
How much energy is required in all to produce a bicycle? The data is scarce and does not distinguish between various types of bicycles. According to an MIT study, the manufacture and maintenance of a bicycle over its entire lifecycle represents 319 MJ of energy per mile traveled (or 199 MJ per kilometer). This type of study assumes that a bicycle has a useful life of 15,000 km. This would amount to a total consumption of 830 kWh of energy. But this figure seems optimistic, or based on the manufacture of very simple bikes.
Another life cycle study of a high-end bicycle estimates its energy bill at 2380 kWh for the frame manufacturing stage alone. Add to that 325 kWh for the wheels and 50 for the chain, not counting the other parts. The total could therefore come to about 3,000 kWh. The average Québec household consumes 72 kWh of electricity per day (compared to 13 in France), so those 3,000 kWh are the equivalent of 41 days of electricity consumption in Québec or 231 days in France. It also adds up to to the equivalent of about 750 kilos of firewood.
In a post-growth world, metal will still be produced in fairly large smelters and aluminum plants in order to minimize energy consumption per unit produced. These factories will supply large areas, at a country or subcontinent scale. However, energy constraints and lower overall needs will mean that these plants will be smaller than the ones we are familiar with because there will be fewer blast furnaces or vessels operating in parallel.
The model of the small machine shop that once met all kinds of local needs for single parts or short series comes to mind. These workshops often had fewer than ten employees, but the model could be extended to 25 or even 50 workers to supply a regional market with bicycles. The production of metals and alloys, on the other hand, was already carried out in large units of several hundred employees as early as the 19th century, mainly for reasons of energy efficiency. It is easier to reach and maintain the necessary heat levels in large crucibles than in small ones. Aluminum will still have to be made in large factories, as the process does not lend itself well to small-scale production. It is also possible, within certain limits, to melt steel and aluminum scrap in arc furnaces. The minimum size of these units ranges from 50 to 100 employees.
Dandelion rubber and tires could be produced regionally. This is currently done on a national and even international scale, but it is not a technical requirement, especially not for relatively simple bicycle tires, as compared, say, to car tires. Regional level production would limit the transportation needs of the various goods and avoid the technical problems linked to the production of a lot of renewable energy at a single large site. When energy is less abundant, decentralized production is easier.
If the option of wooden frames is not long lasting enough, it will be necessary to continue to use extruded tubes. Extrusion presses are large pieces of equipment which do not require many workers to operate, but require a fair amount of energy, which could influence the choice of production sites. This equipment will probably not be employed in bicycle production workshops, but will remain in separate facilities, providing the necessary tubes to manufacturers. The manufacturing of ball bearings will most likely work well on the same model, although it would be easier to integrate it into a production workshop.
Lastly, there is the machining of sprockets, chain wheels and various other parts. These could all be made on the same site alongside the assembly work. Machining is relatively energy efficient and it would be possible for a facility to devote itself to this activity when energy is plentiful and then to assemble the accumulated stock of parts when it gets more scarce. A small regional workshop bringing together all these activities could probably not handle all of them at the same time. Workers could focus on some tasks one day and then shift to some others the next. From a capitalist point of view, this practice does not help to maximize the return on investment. From a degrowth perspective, however, it makes it possible to maintain a reasonable level of energy consumption per unit and to provide work without producing a surplus of bicycles.
A bike designed to be durable and easy to repair can last for decades. There are no figures on the number of existing bicycles in Québec, but based on an estimate of 4.2 million cyclists, we can conclude that there are roughly 5 million bicycles. If they are designed to last 50 years, we would have to replace 2% of the fleet each year, or 100,000 new vehicles annually.
If we assume that 1,500 kWh of energy is needed for the production of a simple post-growth bicycle, the energy requirement of this industry would come to 150 GWh, which is the equivalent of the annual production of about 35 large 2 MW wind turbines operating with a load factor of 25% or, alternatively, of 75,000 tonnes of firewood. This figure includes the energy cost of extracting and processing materials (metal, plastic, rubber, etc.) but leaves aside the question of their environmental impact. Without being unsustainable, the total remains considerable. In addition, the production of a large number of spare parts must be taken into account.
This study intentionally omits two issues that would require additional research. On the one hand, bicycles require roads and streets. Examining the bicycle as a transportation system (rather than an object, as this paper does) would involve measuring the energy cost of roads and making various assumptions about how much they would be used by bicycles (as a percentage) in various scenarios. Bear in mind that, historically, the initial demand for paved roads came from cyclists, who required smooth running surfaces. Cars later benefited from these better roads.
Another interesting line of inquiry would be to ask if we need that many bicycles – or even if we need them at all. The calculations regarding the production of bicycles set out here implicitly assume that the way they are used will remain unchanged. As the required production figures are high, it may be necessary to adopt alternative use patterns or shared ownership of bicycles. We can also assume that a society living more locally would not need to travel as much, which could reduce or eliminate the required bicycle fleet.
This study sought to explore how we might produce bicycles in a post-growth world. It looked at the organization of production, the necessary material and energy resources, and the technical choices required to maintain a sufficient level of productivity without depending on overly complex technologies.
With regard to work organization, we see that several activities can take place in relatively small workshops (less than 50 people), possibly collectively owned by the workers. Some activities, however, such as steel fabrication and extrusion, are not as suitable for small-scale production. They also require equipment costing a few million or even tens of millions of dollars and raising the necessary capital will not be easy for a collective of workers.
When it comes to material resources, none are truly scarce and there are no serious barriers to a sustainable bicycle industry. In a much more energy-scarce post-growth world, however, generating the necessary energy will not necessarily be easy. The choice of production sites could be dictated by the local availability of energy in sufficient quantity, and intermittent production could regulate the production rate of bicycles. This would require significant social adaptation, but is consistent with the degrowth ideal of a slower pace of life.
Technical choices raise some more complex issues. Although the late 19th century was technologically advanced enough to produce bicycles, their purchase price was so high at the time that they were reserved for an elite. We must aim for a cost-price ratio that makes them affordable for the ordinary citizens of a more frugal world to buy them. There is therefore a trade-off to be made between simplicity and a technical efficiency which induces all kinds of dependencies: scarcer resources, longer supply chains, greater capital requirements for the purchase of machine tools.
In short, the big question that arises is just how far to degrow. The example of the bicycle shop shows that there is a tension between the desire for a simpler, more local and more modest life and the desire to preserve a variety of goods and services that make life easier and that extend the capacities of the human body. Giving up too much could make post-growth life miserable. Conversely, too little restraint would leaves the industrial world as ecologically unsustainable as it is now.
This text was initially published in French by Polémos, an independent research group on degrowth based in Montréal, Canada (https://polemos-decroissance.org/). It is released under a Creative Commons 4 license (CC-BY-SA 4.0).
Manifeste du Mouvement québécois pour une décroissance conviviale, https://www.decroissance.qc.ca/manifeste
Bliss, Sam, My Journey Toward Degrowth, https://www.degrowth.info/en/2016/01/my-journey-toward-degrowth
Dubé, Catherine, Prêts pour la décroissance? L’actualité, February 2019. https://lactualite.com/societe/decroissance/
Ezra Bridgman and Quentin Chevalier, Marcher vers la postcroissance, L’échappée belle 3, http://www.lechappeebelle.ca/2015/04/marche-pratique-subversive/
Illich, Ivan, Énergie et équité, Le Seuil, Paris, 1973, p. 49.
Bihouix, Philippe. L’âge des low tech, Le Seuil, Paris, 2014, p.128.
Abraham, Yves-Marie. Guérir du mal de l’infini, Écosociété, Montréal, 2019, p.246.
Polémos. Comment relancer l’économie autrement, Le Devoir, May 22, 2020, https://polemos-decroissance.org/lettre-opinion-refaire-la-societe/
Gaboriau, Philippe. Les trois âges du vélo en France. Vingtième siècle, revue d’histoire, 1991. https://www.persee.fr/doc/xxs_0294-1759_1991_num_29_1_2335#xxs_0294-1759_1991_num_29_1_T1_0021_0000
Institut de la statistique du Québec. Rémunération hebdomadaire et horaire des employés, régions administratives et ensemble du Québec, 2015-2019
Gaboriau, op. cit.
Smil, Vaclav. Energy Transitions. Praeger, Saint Barbara, 2010.
Low Tech Magazine. How much energy does it take (on average) to produce 1 kilogram of the following materials? https://www.lowtechmagazine.com/what-is-the-embodied-energy-of-materials.html
Chevalier, A. Le pissenlit à caoutchouc en URSS. Persée, 1945. https://www.persee.fr/doc/jatba_0370-3681_1945_num_25_275_1811
Dave, Shreya. “Life Cycle Assessment of Transportation Options for Commuters.” Massachusetts Institute of Technology (MIT) February 2010. http://files.meetup.com/1468133/LCAwhitepaper.pdf
Roy, P., Miah, D., Zafar, T. Environmental impacts of bicycle production in Bangladesh: a cradle-to-grave life cycle assessment approach. SN Applied Sciences, 2019. https://doi.org/10.1007/s42452-019-0721-z
Johnson, R., Kodama, A, Willensky, R. The Complete Impact of Bicycle Use: Analyzing the Complete Impact and Initiative of the Bicycle Industry. Duke University, April 2014.
Statistics Canada data quoted in Gagné, Alexis. Une taxe à la surconsommation d’électricité, ou une taxe sur les grandes familles? L’actualité, March 2015. https://lactualite.com/lactualite-affaires/une-taxe-a-la-surconsommation-delectricite-ou-une-taxe-sur-les-grandes-familles
Consommation électricité moyenne en France en 2020: statistiques et analyse, https://prix-elec.com/energie/comprendre/statistiques-consommation-france
Ministère des Transports du Québec, L’état du vélo au Québec en 2015, https://www.transports.gouv.qc.ca/fr/modes-transport-utilises/velo/Documents/etat-velo-2015.pdf
Teaser photo credit: Wooden draisine (around 1820), the first two-wheeler and as such the archetype of the bicycle. By Gun Powder Ma – Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=4406665