Making and breaking links: the transformative potential of shared mobility from a practice theories perspective

Abstract

Shared mobility has the potential to contribute to the transition to a more sustainable mobility system. However, the environmental impacts and the extent of proliferation of the various shared mobility practices differ considerably. It is problematic that the most widespread practice—free-floating carsharing—shows the least environmental potential. Thus, the question arises as to why some shared mobility practices proliferate more readily than others. This paper studies this question from a practice theoretical perspective, focusing on how practices link or do not link with one another. It analyses how various shared mobility practices, as well as the practice of private car travel, connect to other practices via spatial-material and temporal links. The analysis explains why private car travel and, to a lesser degree, free-floating carsharing integrate relatively easily into everyday life, while other forms of shared mobility struggle to do so. This observation leads to the need for far-reaching interventions, both in the making of links of sustainable practices but also in the breaking of links of unsustainable practices. This paper scrutinizes this issue in an anticipatory and theory-based manner and offers suggestions on how to refine practice theoretical concepts regarding inter-practice connections.

Keywords:

• Shared mobility
• practice theories
• sustainable mobility
• connections between practices
• carsharing
• cargo bike-sharing

1. Introduction

To achieve the Paris Agreement’s global climate change goals, countries must drastically change their profoundly unsustainable consumption practices and abandon the ‘imperial mode of living’ (Brand and Wissen 2021). While emissions in some sectors have begun to decrease since 1990, the transport sector’s emissions have consistently increased each year¹ (European Environment Agency 2021). Mobility practices, thus, have a central role to play in this urgent effort to address climate change.

Shared mobility should play an important role in this transformation (European Commission 2020), reducing private car travel and providing mobility solutions in situations poorly suited for public transport, cycling or walking. However, there is disagreement regarding the actual ecological impact of shared mobility. For instance, there is concern that carsharing prevents people from taking public transport (Ceccato, Chicco, and Diana 2021; Nehrke and Loose 2018). While it is difficult to evaluate the ecological impact of shared mobility generally, if we consider specific shared mobility solutions, we can achieve a more nuanced understanding. While free-floating carsharing indeed offers limited ecological benefits, other forms of shared mobility may enable more sustainable mobility practices. The most sustainable solutions, like peer-to-peer carsharing or cargo bike-sharing, have limited appeal, however; free-floating carsharing, by contrast, has become mainstream (Giesel and Nobis 2016). The question arises, then, as to what propels some shared mobility practices into the mainstream but holds others back? This paper examines this question, drawing on the perspective of practice theory.

Practice theories are especially well adapted to illuminate the rise, stabilisation and inertia of mundane social practices. Within the field of sustainable mobility, these theories have proved helpful in analysing both the stability and variability of mobility practices (see for a literature review Kent 2022). As these theories analyse material infrastructure, routines and institutionalization of social practices, they fit well with the central dimensions of mobility.

In this paper, I work with a strand of practice theories revolving around connections between social practices. As mobility practices are ‘connecting practices’ in that they connect the different places in which the practices are performed, understanding these links is crucial. Specifically, I will apply the concepts of temporal and material-spatial links between social practices to different forms of shared mobility (free-floating carsharing, peer-to-peer carsharing and cargo bike-sharing), as well as to the practice of using a private car.

It is crucial to include private car travel in the analysis because too often the research on sustainability innovations like shared mobility focusses only on encouraging the spread of these innovations (Moreau et al. 2020; Hartl et al. 2018; Becker and Rudolf 2018). But for those innovations to take root, we must first understand how current unsustainable practice arrangements define the web of practices into which the innovations must be integrated. This article argues that new practices need not only the forging of links with other practices but also—more importantly—the breaking of links with old practices. An intervention in the form of this making and breaking of links is indispensable; if shared mobility will not be regulated, the most adaptive shared mobility options are likely the ones that will spread, and these are the ones whose environmental value is rather low.

As proposed by Castelo, Schäfer, and Silva (2021), this kind of theory-driven analysis should help to anticipate crucial factors for a transition to more sustainable practices. Hence, one aim of this article is to analyse the prospects for the proliferation of shared mobility practices—and what might hinder this proliferation.

In addition, this article has a second aim at the theoretical level; it discusses and applies concepts of inter-practice connections and suggests further developments regarding rules and regulations as well as the breaking of links between practices.

To achieve both aims, the article begins by discussing the potential environmental benefits of shared mobility schemes. Then, in the third section, the theoretical perspective is introduced and applied to the practice of private car travel and, in the fourth, to shared mobility practices. The fifth section describes the consequences of these applications for both aims; it is divided into ‘practical reflections’ and ‘theoretical refinements’. The sixth section concludes the article.

2. Shared mobility from an environmental perspective

The transport sector is highly problematic in terms of carbon emissions (European Environment Agency 2021), and the dominance of private cars among options for individual mobility is among the main causes. One reason why the private car maintains supremacy is that it can go almost anywhere; neither public transport, cycling, nor walking can match the private car’s ability to reach most destinations. Thus, a non-car-centric mobility system must be multi-modal—rooted in sustainable modes of transportation, like public transport or cycling, but supplemented by forms of shared mobility to access destinations otherwise unreachable by non-car-based mobility options (Riegler et al. 2016, 77; Ruhrort 2020). Shared mobility is an umbrella term for services that provide short-term access to de-centrally allocated vehicles. These services comprise a variety of organisational models (station-based, free-floating, hybrid), are run by different actors (private companies, public entities, civil society organisations, hybrid constellations) and exist for various vehicles (cars, bikes, cargo bikes, e-scooters, mopeds). Lately, mobility-as-a-service systems are gaining importance. These systems bundle information about different mobility options, like public transport, shared mobility and taxi services, in one web application, including—ideally—reservation and payment modalities.

It has been suggested that the number of cars tolerable in a sustainable multi-modal system must decrease from 485 cars per 1000 inhabitants to 150 cars per 1000 inhabitants (Umweltbundesamt 2017). Against this backdrop, shared mobility seems a promising option for the transformation of individual mobility (Shaheen, Chan, and Gaynor 2016; Witzke 2016), and studies could affirm environmental benefits like reduced car ownership for carsharing (see for an overview Kent 2014).

Critics claim, however, that shared mobility options, particularly carsharing, might have negative environmental effects that outweigh the positive ones—for example, if carsharing were to replace public transport (Ceccato, Chicco, and Diana 2021; Schwedes 2021). There is also concern that carsharing might strengthen the car-driving habit and prop up the automobile industry by extending its market power (Giesel and Nobis 2016; Riegler et al. 2016).

Effective ecological performance depends on whether shared mobility can bring about changes in car ownership and in the distance travelled by car (Ruhrort 2019). Thus, the effects of shared mobility will be positive only if the distance travelled by car decreases and a modal split develops whereby more sustainable modes of transportation, like public transport, cycling, or walking, flourish. Depending on the organisational nature of the shared mobility schemes, empirical findings show diverging results related to those two determining factors (Gossen, Pentzien, and Peuckert 2019). In addition to those two determining factors, the extent to which the shared mobility options proliferate is crucial, as high user numbers are needed for a significant reduction in the total number of cars.

Therefore, to estimate the transformative potential of shared mobility, it is important to differentiate among options. For this endeavour, this article studies three very different forms of shared mobility: free-floating carsharing, peer-to-peer carsharing and cargo bike-sharing. These were chosen to ensure inclusion of both more widespread and less popular mobility options as well as options with a range of environmental potential. These forms can be defined as follows:

2.1. Free-floating carsharing

Free-floating carsharing schemes are run by private companies. With free-floating or one-way carsharing, users begin and end their trip at different locations within a defined area. This contrasts with station-based roundtrip or station-based one-way carsharing, in which users return the car to a defined place—in the case of round-trip carsharing, to the same location where they got it, and in the case of one-way carsharing, at a different location. An example of this form is Share Now.

2.2. Peer-to-peer carsharing

With this form of station-based carsharing, privately owned cars are shared. Usually, this occurs in a station-based round-trip manner. In most cases, it is organised through a digital platform that connects car holders and lessees. These platforms are run by civil society organisations or private companies. Examples are Getaround, or Seestadt Mobil.

2.3. Cargo bike-sharing

Cargo bike-sharing is most commonly organised by civil society organisations in a station-based round-trip manner. The shared bikes are owned by individuals or a community. Public entities and a few private companies run a minority of cargo bike-sharing schemes. Examples are milla.bike or Lastenradkollektiv.

In terms of user numbers, free-floating schemes are by far the most widespread (Münzel et al. 2020), and because of their proliferation, carsharing is sometimes presented as a mainstream phenomenon. Regarding the environmental impact, there are considerable differences among the variants; station-based carsharing has been shown to replace private car ownership at a higher rate than free-floating carsharing (Ruhrort 2019; Nehrke and Loose 2018; Ludmann 2019). This is reflected in car ownership rates: slightly more than half of free-floating carsharing users own a car or live in a household with a car, but this number drops to around 25% for station-based carsharing users (Riegler et al. 2016). Cargo bike-sharing is not yet well researched, but one study found that almost no users of a cargo bike-sharing scheme used a car regularly (Becker and Rudolf 2018). The potential for cargo bikes to cut mobility-induced greenhouse gas emissions might be simultaneously high and underestimated: cargo bikes could help reduce the number of under-5-kilometre car trips that require cargo capacity. E-scooter or bicycle-sharing schemes are not included in the analysis, as studies could not find any significant environmental benefits related to those forms of shared mobility (Becker and Rudolf 2018; Moreau et al. 2020).

Thus, the most widespread form of shared mobility, free-floating carsharing, shows only low ecological impact. Scholars warn, however, against categorizing shared mobility options as environmentally ‘good’ or ‘bad’. They emphasise that any assault on the current unsustainable mobility system helps to break with extensive car use and ownership (Ruhrort 2019, 2020). While this argument is valid, if the long-term environmental impact is to be significant, forms other than free-floating carsharing must enter the mainstream, too.

Therefore, to capitalize on the potential of shared mobility, we need to understand why some mobility practices become mainstream and others do not. The next section will introduce the theoretical perspective used for this endeavour.

3. A practice theoretical understanding of private car travel

3.1. Key terms of practice theories

Traditionally, theoretical approaches to understanding and changing consumer behaviour in the field of sustainable consumption have been based on methodological individualism. In those theories, individual behaviours form the basis of social structure, and they assume a rational, deliberate, conscious, autonomous consumer. They assume that changing the knowledge, attitudes and values of individuals also changes consumption choices. In transport planning, approaches from this theoretical family—like rational choice theory or the socio-psychological theory of planned behaviour and its developments—are still frequently used (Kent 2022).

However, the assumption of a rational, profit-maximising consumer does not hold; findings show that despite the considerably greater expenses associated with car travel than other forms of mobility, it remains the dominant mode (Andor et al. 2020). Likewise, regarding shared mobility, it is not the least costly option—for example, many cargo bike-sharing schemes are free of charge—that proliferate most widely.

In an attempt to overcome the limits of both behavioural theory and systemic perspectives, a wave of practice-oriented social theories has emerged over the last two decades (Reckwitz 2002; Schatzki, Knorr-Cetina, and Savigny 2001; Warde 2005; Shove, Pantzar, and Watson 2012; Schatzki 2002). These theories assert that there is no such thing as a responsible, rational consumer. Instead, a major part of everyday life consists of routinized social practices that are firmly anchored in materials, infrastructures, rules, norms, cultural understandings, conventions, symbolic meanings and incorporated knowledge. Practice theories shift the focal point of an analysis away not only from individuals but also structural entities (like ‘the market’ or ‘the state’), positioning social practices as the central units of analysis. These theories assume that there are broadly shared agreements concerning the ‘normal’ way to do something, and recalibrating these normalities involves more than changes in value systems or adapted economic incentives (Reckwitz 2002; Schatzki, Knorr-Cetina, and Savigny 2001; Warde 2005; Shove, Pantzar, and Watson 2012; Schatzki 2002). Shove, Pantzar, and Watson (2012) defined ‘practices’ as specific connections between materials like things, technologies and infrastructures; competences like knowledge, skills and understanding of how to do things; and meanings, like worldviews, aspirations and connotations. Practices appear as ‘practice-as-entities’ (abstract and idealised forms of the practice) and ‘practice-as-performances’ (the enactment of the practice, the doing) (Schatzki 2002).

3.2. The practice of private car travel

What does such a perspective mean for the understanding of mobility practices? How can the practice of private car travel be framed using practice theory? Although there are many variations of the practice of private car travel, Shove, Pantzar, and Watson (2012) hold that it is standardised enough to speak of a recognizable entity. Private car travel involves materials like the road system, the car, the petrol distribution system and parking sites; skills like how to drive a car, knowledge of traffic rules, car maintenance and insurance; and meanings like freedom, flexibility, independence, convenience and masculinity. Through the repeated use of these materials, the repeated applications of these skills and the repeated ascriptions of such meanings in a ‘circuit of reproduction’ (Shove, Pantzar, and Watson 2012), the practice persists. Importantly, this process is never stable, but in constant flux. If this constant reproduction becomes disrupted, or if one of its basic practice elements is not continuously reproduced by its carriers (individuals), the practice either loses meaning and disappears (‘ex-practice’) or is replaced by a new or modified practice (‘proto-practice’) (ibid.). Historically, in the immediate post-war era (the 1950s), the practice of cycling lost its meaning and material basis as a safe and reasonably fast mobility practice and was replaced eventually by the practice of car travel. The loss of these meanings was related to the development of a ‘system of automobility’ (Urry 2004) that reconfigured space and distances between the locations of different practices—the material element—in favour of car travel, set new standards for ‘safe and fast’ mobility and thereby marginalised the practice of cycling (Watson 2012).

3.3. Conceptualising connections between practices

Practice theories have been successfully used to understand the transition to sustainable personal transport (see for an overview Kent 2022). However, the literature to date is ‘missing the importance of practice bundles’ (Kent 2022, 231). Indeed, mobility practices are by definition ‘connecting practices’, and it is clear that they cannot be fully understood in isolation (Shove and Spurling 2013; Shove, Watson, and Spurling 2015; Castelo, Schäfer, and Silva 2021; Watson 2012). Mobility practices do bring people from the place of enactment of one activity (e.g., work location) to the place of enactment of another activity (e.g., shopping mall), and in so doing interrelate with many practices. Scholars have used different terms to describe this close relationship: mobility practices and end use practices (Spurling 2020), direct transport practices and practices facilitated by transport (Kent 2022), trips and the activities enabled by them (Watson 2012). Mobility studies speak of ‘derived demand’ (Bamford 2001) in this regard. But such an understanding should not assume a fixed demand to which mobility practices passively cater. Instead, connected practices continuously interact with and shape each other (Spurling 2020) dynamically, creating contingent practices. For example, suburban housing practices and car travel developed together, creating ever more ‘demand’ for car travel.

Only recently, work in the field of practice theories has explicitly focused on links between practices² (Hui, Schatzki, and Shove 2017; Shove, Watson, and Spurling 2015; Bellotti and Mora 2016; Castelo, Schäfer, and Silva 2021) and defined temporal as well as material-spatial dimensions of connections as essential (Blue and Spurling 2017). In the following, I discuss some thoughts on these links.

3.3.1. Material-spatial links

Hui, Schatzki, and Shove (2017) elaborated on the term ‘threading through’ to allude to things that move through and link the nexus of practices. This idea of materials that interweave practices was present in early writings by Schatzki, Knorr-Cetina, and Savigny (2001) and further elaborated upon by Hui, Schatzki, and Shove (2017). Schatzki (2010) developed the idea of material arrangements ‘prefiguring’ social practices, arguing that some linkages between practices and material arrangements affect future linkages, by making them easy or hard, tiresome or invigorating. The (non-)existence of a good cycling infrastructure, one that includes cycle superhighways or broad bike lanes that can accommodate cargo bikes, is a prime example of how tiresome or easy it is to use the bike.

In parallel to the concept of ‘threading through’, the concept of ‘suffusing’ was developed (Hui, Schatzki, and Shove 2017). It also refers to the way elements of practices connect with other practices, but it refers to less concrete and distinct trajectories. Whereas elements that ‘thread through’ follow identifiable paths, elements that ‘suffuse’ spread in a more ethereal, more liquid, less tangible way (Hui, Schatzki, and Shove 2017). Thus, the concept of ‘suffusing’ could be productively applied to the way meanings like freedom, masculinity, independence, and so on travel across practices. This article focusses on the ‘threading through’ of material-spatial links, but as meanings play a very important role in mobility practices, working with the concept of ‘suffusing’ seems promising.

3.3.2. Temporal links

Turning to temporal relations, Shove, Pantzar, and Watson state, ‘enacting a practice is a matter of weaving it into an existing rhythm and of honouring temporal injunctions inscribed in concepts of proper performance’ (Shove, Pantzar, and Watson 2012, 127).

While there is broad agreement in practice-theories that temporal features such as periodicity, sequence, synchronisation, duration and tempo are important, the role of time and its relation to practices is disputed. Southerton (2003) understands time as existing within practices, whereas Shove, Pantzar, and Watson (2012) understand it the other way around, seeing time as a temporal landscape that structures practices. Trying to overcome these dualistic perceptions, Blue (2019) develops a recursive understanding of practices and temporalities and presents the interesting concept of ‘entrainment’. This article goes in its understanding of time in line with Blue, and his concept of ‘entrainment’ (how specific practices set the pace for others) could be productively applied in a further elaboration.

Central concepts regarding temporal aspects of practices are sequence and synchronisation, as ways of ‘inter-locking’ practices (Blue; Shove and Spurling 2013; Blue, Shove, and Forman 2020; Spurling and McMeekin 2015; Blue 2019). A sequence of practices is defined as the way practices precede or follow each other. While some sequences are fixed, others offer room for variation and hence potential to reorganise. One such example is to reorganise the sequence ‘shower at home in the morning, drive the car to work, start working’ into the sequence of ‘cycle to work, shower at work, start working’. Empirical examples have shown changing material environments, like installing showers at the workplace, could indeed reorganise the sequence and attract practitioners for cycling (Buehler 2012).

Thinking about sequences also offers the possibility of thinking from the perspective of avoiding mobility. This is crucial as the strategies of ‘improving’ and ‘shifting’ must be accompanied by the strategy of ‘avoiding’ mobility to reach climate goals (Bongardt et al. 2013; Rammler 2016). If ‘what comes after’ private car travel is not location-bound, mobility practices can be avoided completely. Avoiding environmentally harmful mobility by replacing some work-related physical mobility with virtual communication has long been discussed as a potentially productive way to reduce overall car travel (Spurling and McMeekin 2015; Watson 2012; Rammler 2016; Spurling 2020). Against this backdrop, it is of significant interest how the immense expansion of videoconferencing and remote work induced by the COVID-19 pandemic will reorganise the links between work and travel. Will it help to reduce car travel or lead to rebound-effects, as people might tolerate longer commutes if they do not always need to be present at the workplace?

To illustrate the concepts of material-spatial and temporal links and highlight their analytical potential, I will apply them to the social practice of private car travel.

3.4. Private car travel and its inter-practice connections

In material-spatial terms, the road system and parking sites are the primary infrastructures ‘threading through’ diverse practices, and these connections smoothly integrate the practice of using the private car into the web of everyday life. To illustrate this, the intimately linked practices of grocery shopping and private car use serve as an example. An essential feature of this linkage is the ability to park the car close to one’s house or apartment and in front of the—oftentimes out-of-town—grocery store. This ability necessitates the dedication of large parts of public space as parking sites. Indeed, it is the broad availability of parking sites that allows the car to serve as the normal means of transportation (Ruhrort 2019; Spurling 2019) and, in Schatzkìs terms, ‘prefigures’ car use (Schatzki 2010; Schatzki 2017). Also, the material making of some widespread housing practices strongly prefigures car travel. Examples are the material arrangements of scattered single-family houses, urban sprawl, and the concept of the functional city that divides housing, working and recreational practices into separate areas. The proliferation of these forms of housing and spatial planning co-evolved with the massive motorization of the population in the post-war era, resulting in practices that are now mutually dependent and hence very hard to change (Urry 2004). Fiscal regulations that encourage car travel (like tax relief for car commuters) and serve the economic interests of the automobile and oil industry further sustain the ‘system of automobility’ (Manderscheid 2021).

The connectivity of car travel is high regarding temporal links, as well. Car travel connects with all sorts of activities and fits into many kinds of sequences; some of these fit especially well together, and some even seem designed specifically to make use of a private car. For example, many sports activities require private equipment that needs to be carried by car, and many forms of holiday travel depend on the possibility of bringing many belongings (Manderscheid 2019). Mattioli, Anable, and Vrotsou (2016) found that practices of escorting children, shopping and carrying heavy goods are the most ‘car-dependent’. For many other sequences, it is less obvious why a private car usually connects the sequence, but many car owners are rather monomodal (Ruhrort 2019). When using a private car, the sequence usually involves just a few elements and a short duration: the decision to go somewhere can be followed immediately by starting the trip, especially if the car is parked only a few steps away, because car-centred infrastructure threads through many practices.

Thus, social practices show intra-practice as well as inter-practice links; practices hang together internally through the connections of their elements, but also externally by relations between different practices. Furthermore, these two levels hang together, as the points of intersection of both kinds of connections constitute the practice’s elements. ‘In our account, elements simultaneously figure as ingredients of practices and points of connection between them’ (Shove and Trentmann 2019; Shove, Pantzar, and Watson 2012, 122). So, changes in one practice might induce changes in connected practices (Spurling and McMeekin 2015; Watson 2012). These theoretical reflections about how practices change directly lead to a central question in the context of sustainability: the question about the steerability of practices.

Before applying the concept of temporal and material-spatial links to shared mobility practices, I will briefly discuss practice theoretical perspectives on interventions and how this article positions itself in this regard.

3.5. Interventions in practices

Most practice-theoretical scholars will agree that theories based on methodological individualism strongly inform current policymaking and lead to ‘soft’ policy instruments, like information campaigns, publicity, price signals and the support of technological innovations. However, based on the key assumptions of practice theories, such interventions are doomed to fail (Watson 2012; Kent 2022; Shove 2014).

Despite this agreement, the issue of interventions is contested in practice theories and leads to two opposing positions. One position depends on viewing social change as a product of the fundamentally uncontrollable evolutionary dynamic of social practices (Brand 2011, 190). The other position views practices, their elements and their connections as promising entry points for social change; in this sense, practice theories can deliver important insights for policy measures. This article embraces the second position, but with caveats. Unlike some scholarship (Vihalemm 2015), it does not suggest that practice theory can be easily applied to the design of social change programs but instead argues that practice theory analysis usually demonstrates the opposite—that practices are hard to change because they are embedded within material infrastructures, dominant belief systems, far-reaching norms, and regulations and cultural conventions (Spurling and McMeekin 2015; Shove 2014). A position informed by practice theory also clearly contrasts with behavioural economics positions and the related policy instrument of ‘nudging’. While each position focusses on routines, behavioural economics focus on individual choices and behaviours, detached from societal constraints. Nudging, therefore, falls within the framework of methodological individualism. Practice theories, by contrast, underline exactly those societal constraints and hence stand in opposition to behavioural economics (Reid and Ellsworth-Krebs 2019).

In this vein, this article emphasises that regulations, i.e. rules imposed by public authorities, must go beyond the ‘soft’ policy instruments cited above and embrace instruments that restrict or eliminate choices; pull-measures and the making of links cannot work without push-measures and the breaking of links between practices. The position embraced by this work parallels the work of Watson (Watson 2012) and Spurling and McMeekin (Spurling and McMeekin 2015). Those contributions highlight the radicalism of changes necessary to decarbonize mobility practices and strongly suggest that the analysis of how practices hang together, or ‘interlock’ (Spurling and McMeekin 2015) is especially germane for this work.

4. Shared mobility practices and their inter-practice links

Applying the concepts of material-spatial and temporal links to the practice of private car travel shows that the practice connects material-spatially as well as temporally to many other social practices in multiple ways. How do shared mobility practices perform in this regard, and what does the application of the same theoretical framework reveal about their connectivity?

4.1. Temporal links

Examining different forms of shared mobility reveals significant differences in the sequences they are involved in. Peer-to-peer carsharing is often used for weekend trips to rural areas or the transportation of heavy goods, and its use is usually planned well in advance, with a use duration of at least a couple of hours (Svennevik, Julsrud, and Farstad 2020). Free-floating cars, by contrast, are mostly used for spontaneous drives in the evening hours, for quite short distances and half of these drives are shorter than 5 kilometres (Riegler et al. 2016). For those purposes, free-floating schemes are especially well-adapted, as the price is calculated per minute and the operating area is limited to urban areas (returning the car outside the urban operating area is not allowed). Shared cargo bikes are typically used for inner-city travel that involves considerable transport capacity. Drinks, food, furniture and children are among the most frequently transported things or people (Becker and Rudolf 2018). Thus, the shared mobility options seem to connect with some practices more than with others. This is the reason for the diverging environmental impacts discussed in the second section.

The question that emerges now is why it is difficult to integrate into everyday life the shared mobility practices that replace private cars and fit within car-dependent sequences. A more detailed understanding of sequences—one that analyses how practices and their individual steps expand through the duration of their performance– can deliver insights in this regard.

Depending on the specific practice, the number of elements involved in the sequence and their ‘temporal stretch’ (as I would call it) varies considerably. For free-floating schemes, the duration of the sequence is rather short, because free-floating schemes utilize on-demand booking, and insurance and payment do not need confirmation for every use. Peer-to-peer carsharing, by contrast, requires more planning. The vehicle may need to be booked days in advance, the time and place for the handover of the key must be arranged, and usually the booking request must be validated by the owner of the car, which adds one more step to the sequence of the practice. In many cases, the insurance must be verified for each trip, and often the car or bike is parked a considerable distance away. The beginning of the sequence (the booking request) might be far removed in time from the mobility practice itself due to the arrangements that need to be made. Additionally, the sequences must be planned through to the end, in the sense that the return time must be indicated in advance. With cargo bike-sharing, the sequence involves the same elements as peer-to-peer carsharing but can run even longer if new users need instructions on how to use the bikes.

While all forms of shared mobility require checking the availability of the means of transportation, booking, and payment, the duration of each differs considerably. Only free-floating carsharing involves sequences that are comparably as short as private car travel. If the temporal flexibility and rapidity of private car travel serve as the norm, the comparable duration of free-floating carsharing might be the key to its successful proliferation.

4.2. Material-spatial links

Viewing shared mobility practices through the lens of material-spatial links offers important insights.

At first glance, the infrastructure needed for carsharing—like roads, petrol stations and parking sites—is the same as for private car use (Kent and Dowling 2013). But the importance of this car-centred infrastructure varies considerably among the different shared mobility schemes: while peer-to-peer carsharing users demonstrate a multi-modal split, relying heavily on public transport and cycling infrastructure as well, free-floating carsharing users are less multimodal; they often combine carsharing with private car use (Ruhrort 2019). This means that free-floating carsharing can integrate well in the currently dominant car-centred infrastructure and depends less on other infrastructures. Peer-to-peer carsharing, instead, indirectly depends more on public transport and cycling infrastructure. In the case of cargo bike-sharing, the necessary infrastructure, like wide bike lanes and appropriate bike parking in dense urban areas, is often poor. Thus, the infrastructural requirements are high for cargo bike-sharing because not only is an adapted bike infrastructure necessary, but also a good public transport infrastructure. If these infrastructures are not provided where connected practices occur, like working places, shopping areas, educational institutions, sport venues, friends’ houses, etc., then it will be hard for the practice of using a cargo bike to connect with those other practices.

Regarding shared mobility, secondary to the importance of material infrastructure is materials in the sense of devices that provide points of intersection with other practices. For some materials, we can observe how they and their related competences thread through different practices. Users book mobility services ever more frequently on their smartphones. For some mobility services, the smartphone is the only way to access them (e.g. for most e-scooter sharing and free-floating carsharing schemes). They are needed to locate the vehicles via GPS and to use the keyless opening feature. Inherently linked to those apps are the competences to use them. Those competences migrate between practices; the apps of the most widespread free-floating carsharing schemes integrate elements that practitioners likely know from other websites or apps. An example is the payment method; a common way to pay for shared vehicles is to use a PayPal account or credit card that is already connected to the user’s smartphone for online shopping. Peer-to-peer carsharing and cargo bike-sharing, instead, usually require in-person key delivery involving the owner and the borrower. This not only adds an element, and thus time, to the sequence of these practices, but it also does not link or build bridges to other practices. The key to the peer-to-peer shared car or bike is not ‘already there’ like the smartphone; it is a distinct material element of the practices, unrelated to other practices, that needs to be integrated. It seems—as was found by Kent and Dowling on carsharing (2013)—that the fewer additional materials, devices and related skills to be integrated in the web of everyday life, the easier it is for new practices to stabilise.

Materials that require the physical meeting of people seem emphatically to block the proliferation of shared mobility practices; the often-cited motto ‘sharing is caring’, the idea that the sharing economy creates social bonds, is questionable. Based on user statistics and the lengths sharing schemes go to eliminate the need for in-person meetings, it seems that the fewer social interactions are involved, the easier it is to integrate and grow the practice. Indeed, sharing practices based on social interactions seem to inhibit mainstream uptake of the practice. A study on social innovations for sustainable consumption confirms this finding; it concluded that alternative consumption practices, like carsharing, diffuse more widely if they are less innovative, less community-oriented and rooted in personal relationships, less dependent on self-organisation and more formalised (Jaeger-Erben, Rückert-John, and Schäfer 2017).

Viewing shared mobility practices through the analytical lens of connections between practices shows that some shared mobility practices are difficult to integrate temporally into everyday life because of their extended duration. This sharply contrasts with the practice of private car travel, which is well adapted for the accelerated pace of everyday life. Moreover, in material-spatial terms, private car travel seamlessly connects with many other practices. In the case of shared mobility and multi-modal mobility patterns, these connections are more fragile and contested, or in some cases non-existent. The next section offers some practical considerations based on these theoretical reflections.

5. Making and breaking links: practical reflections and theoretical refinements

5.1. Practical reflections

Scrutinising the sequences of different forms of shared mobility shows that it is harder to integrate peer-to-peer carsharing and cargo bike-sharing into everyday life than free-floating carsharing, a finding in accord with Kent and Dowling (2018), who found that issues related to temporal flexibility pose big challenges to carsharing practices, and with (Priddat 2015) who stated that community-based sharing models require a different and more time-sensitive organisation of everyday life.

Given these observations, there seem to be two possible paths to supporting the uptake of sustainable shared mobility practices like peer-to-peer carsharing or cargo bike-sharing; either reduce the temporal duration, the need for planning and the length of the sequences of those practices (thus aligning them with more proliferated shared mobility practices and encouraging them by making links) or adjust the temporal structure of everyday life to accommodate the more sustainable practices. The first option could be accomplished, for example, by supporting carsharing within residential buildings, which would likely reduce the temporal duration of the sequence. It is also possible, and probably more effective, to come from the other direction—breaking links—by extending the duration associated with the practice of private car use, for example, by reducing parking spaces or creating car-free zones, thus prolonging the sequence by prolonging the time needed to park and reach the car. To compensate for the higher connectivity of free-floating carsharing compared to peer-to-peer or cargo bike-sharing, regulatory interventions could help to balance the starting conditions; these interventions could include more support for peer-to-peer schemes and cargo bike-sharing than for free-floating ones; little support for fleets that consist of small cars with little cargo capacity; influencing free-floating schemes to expand their operating area to suburban areas and to include incentives for longer distances or durations; reductions in the cost for shared mobility for people holding season tickets for public transport. Such regulations could help sustainable shared mobility practices forge more links with other social practices.

The second option would be to adjust time structures that seem more conducive to fast practices and short sequences (Smetschka et al. 2019). Peer-to-peer carsharing or cargo bike-sharing, like other sufficiency oriented practices, involve planning in advance Might the competences of planning in advance and time availability be crucial to realise more sustainable everyday practices, especially those directed towards sufficiency? If yes, the chances might be low for those alternative mobility practices to proliferate given the increasing societal acceleration. Further reflections on this issue might involve the topic of working-time reduction; however, the potential rebound effects of working-time reduction may hinder the realisation of environmental benefits (Reisch and Bietz 2014). Additionally, temporal flexibilisation, digitalisation and increased remote working could enable the proliferation of more time-consuming practices and the avoidance of car travel. Like the question of working-time reduction, these issues are also complicated from a social perspective, but as they might offer opportunities more powerful than moral pleas to reduce private consumption for an urgently needed sufficiency, they should be considered. What becomes clear is that mobility is influenced by policies other than those directly focussed on transport; for example, policies related to education and employment also play an important role (Watson 2012). Thus, the cross-sectoral nature of sustainable mobility renders policy-makng particularly complicated (Shove 2014).

5.1.1. Devices and applications that thread through

In terms of material-spatial connections, examining the devices involved in the sharing process helps to identify ways to support the making of connections with other practices. The less the need to integrate material devices into the practice sequences, the easier the uptake of a shared mobility practice. Therefore, support for keyless opening systems in peer-to-peer carsharing can help in making connections with other practices. This innovation would require financial support for the respective appliances or ‘smartcar-software’ that is introduced to the market, works across car brands and replaces car computers. If sharing instead of owning cars were to become the new normal, standard installation of such software could help peer-to-peer carsharing create links with more practices. In fact, some peer-to-peer carsharing schemes run by commercial platforms (e.g. Getaround), as well as the mainstreamed accommodation sharing platform Airbnb, invested in offering keyless opening and immediate booking systems, with no need to wait for approval of the car or flat owner or a meeting between the owner and renter.

Another way to support the integration of shared mobility practices is to support mobility-as-a-service systems. With these systems, the digital platform threads through many different mobility practices and thus increases their connectivity. Currently, public entities like public transport providers and private companies (e.g. Free Now) are pushing mobility-as-a-service systems. It is unclear which entity will play a more important role in the future, but thinking about that eventual dominance is of utmost importance, since public and private actors operate under completely different logics; whereas public actors might want to integrate more sustainable forms of shared mobility in mobility-as-a-service systems, for private actors, such considerations might be of no importance compared to easily integrated but not necessarily sustainable shared mobility options. In this context, regulation of data sharing processes is crucial. Currently, EU law demands that public transport providers share their real-time data as open data—but the same does not apply to private companies. This puts public actors potentially at a disadvantage. Regulations at this level might be important for the connectivity of the different shared mobility practices (Piétron, Ruhaak, and Niebler 2021). Also, investing in a good public transport and cycling infrastructure can support the diffusion of shared mobility—especially forms other than free-floating carsharing.

5.1.2. Breaking unsustainable links

Mobility studies hold that even more important than support for the making of links is the breaking of material links underpinning unsustainable mobility practices because those are key to maintaining them (Ruhrort 2019; Manderscheid 2021). Ruhrort showed, for the German context, how car-centred infrastructures and regulations continuously reproduce the practice of private car travel. For example, traffic legislation declares that the function of a street is primarily to provide space for motorised private transport. Suchregulations lead to the conversion of public space for comparatively low cost into parking sites, or to requirements that homebuilders finance parking sites (Ruhrort 2019). If the material needs associated with the practice of driving—like space for parking wherever one goes—were not part of the material making of so many other practices, the practice could not be sustained (ibid.). Intervening in those linkages, breaking some connections, and unhinging the practice of private car use from its close connection with several other practices is crucial for a mobility transformation.

This short reflection on policy measures includes pull measures supporting sustainable shared mobility practices as well as push measures making private car travel less attractive, suggesting not only how to make links between shared mobility and other practices, but, more importantly, also how to break links. I argue that the chances for more sustainable mobility practices to spread and integrate into everyday life will only be sufficiently high if unsustainable practices ‘make room’, losing their dominant role. The discussion showed that these processes of (dis)integration often involve regulatory interventions. Therefore, I conclude that regulatory interventions, especially such ones that break links between practice bundles involving unsustainable forms of mobility, are key for a transition towards a more sustainable multi-modal mobility system. However, in practice theories, the role of regulations and processes of breaking links are not described in a satisfying manner. In the following, I will briefly present practice theoretical conceptualisations of regulations and the breaking of links and introduce some suggestions for further development.

5.2. Theoretical refinements

5.2.1. Rules and regulations in practice theories

Although interventions are a focus of practice theories, the conceptual understanding of regulations—or, on a more general level, the understanding of rules—is not very well elaborated. Watson himself observes that, in the work of Shove, Pantzar, and Watson (2012), rules are not well addressed, at least not explicitly (Watson 2017, 172). Schatzki is the theorist who gives rules the most prominent role. He understands rules as one of four elements that link the doings and sayings of a specific practice, and by rules he means ‘explicit formulations, principles, precepts, and instructions that enjoin, direct, or remonstrate people to perform specific actions’ (Schatzki 2002, 79). Further, he applied the idea of ‘threading through’, later on applied to materials, to rules. He states that rules can be an organizational component that ‘cuts through’, crisscrosses and interweaves practices (Schatzki 2002, 87). For the understanding of mobility practices, I hold that it would be useful if rules and regulations be given more importance in theoretical accounts. The concept of ‘threading through’ can convincingly be extended to rules and regulations, as Schatzki did in his early writings. If rules and regulations were positioned in the element-model developed by Shove, Pantzar and Watson, it would pair well with the element of competences; how to do things is shaped, among others, by formal rules. In the case of car driving, this is especially evident, as driving competences are institutionalised by the driving license that represents the knowledge of formal rules. Also, a better understanding of rules could help practice theories address the criticism that they are blind to questions of power (Watson 2017).

5.2.2. Breaking inter-practice links

Although it is often emphasised that a transformation of practices must include the breaking of links between practices (Shove, Pantzar, and Watson 2012; Blue et al. 2016; Shove 2014) and that ‘governments have a hand in forging and breaking some of the links involved in the surprisingly uncontrollable, surprisingly living system that is daily life’ (Shove 2014, 426), a more detailed description of how links between practices break or are broken is rare. I suggest that one way to break links between practices is to target the connecting elements, those that thread through different practices—not only materials and infrastructures but also regulations threading through practices. An example is the smoking ban that managed to break the link between smoking and eating at a restaurant (Blue et al. 2016).

Another way to break links between practices could be to interfere in sequences and alter the logic by which they cohere. An especially important question is whether reaching a location is necessary in the first place. Practice theories insist that there is no given ‘demand’ for mobility practices; instead, it is a matter of how the coherence of mundane social practices constitutes the organisation of social life (Watson 2012; Spurling and McMeekin 2015; Shove, Watson, and Spurling 2015). The COVID-19 pandemic provides a prime example of this contingency, showing the fragility and assumed nature of temporal and spatial structures of everyday life. It showed that widespread routines like commuting to the workplace on a daily basis can change if the sequences they integrate into change.

Breaking links might be a powerful instrument, as it changes the environment into which mobility practices must integrate and might make private care use more difficult to integrate into everyday life. Thus, I argue that only by accompanying the creation of links with the breaking of links, integration with disintegration, pull measures with push measures, can the oft-cited upscaling of social innovation take place. Therefore, a thorough theoretical understanding of breaking inter-practice links is needed.

6. Conclusion

A key message of this article—one shared by Loske (2018) regarding the Sharing Economy in general—is that (shared) mobility practices must be regulated if they are to aid a transition to a more sustainable mobility system. If not, the most adaptive shared mobility options will likely diffuse—and these are those whose environmental value is relatively low.

This conclusion is based on the application of practice theories focusing on intra-practice connections to the field of shared mobility. This paper analysed how free-floating and peer-to-peer carsharing, cargo bike-sharing and private car travel link to other practices outside the realm of mobility and integrate into the nexus of everyday life. In summary, from a practice theoretical perspective, it is not surprising that free-floating carsharing—the form of shared mobility with the lowest environmental potential—has the highest user numbers and integrates more easily in everyday life. Current car-centred spatial planning accommodates free-floating carsharing more easily than other forms of sustainable mobility. Peer-to-peer carsharing as well as cargo bike-sharing involve comparatively longer sequences that require more time-sensitive organisation of everyday life and are not well compatible with accelerated societies.

Hence, there are specific practice arrangements and links between practices that hinder the proliferation of the environmentally more promising shared mobility practices. In many cases, the space for integration of promising practices is limited because private car use and ownership connects with so many other tightly interconnected practice complexes. Interfering in these unsustainable constellations by making links, but more importantly by breaking them, will be necessary for shared mobility to have any significant positive environmental impacts. Otherwise, there is little chance that alternative mobility practices can compete with the systematically privileged practice of using a private car (Canzler 2021; Ruhrort 2020). While the making of links—supporting shared mobility practices in integrating in everyday life—is a common strategy in research and in politics, the breaking of links is often neglected; there is literature on diffusing, spreading, upscaling and mainstreaming sustainable mobility practices and innovations, but little on how this is related to the ex-novation of the currently dominant system of automobility. However, breaking links changes the starting conditions and checks the advantages of unsustainable mobility practices. At least in the field of mobility practices, ex-novation might be the central ingredient for successful innovations. Thus, the proliferation of sustainable shared mobility practices necessitates the ex-novation, the radical deprivileging of private car travel. Clearly, the making of links is politically more comfortable and aligns with neoliberal positions—but the focus on that strategy hinders a more effective transformation of the mobility system.

Additionally, it is important to note that intervening in links between mobility and other practices creates the need for challenging cross-sectoral policy-making. Considering the sector of spatial planning is quite common in mobility studies, but when it comes to shared mobility, the importance of spatial planning is often overlooked. Less common is the reference to what is known as ‘time politics’, and I agree with Henckel (2014) that it needs much more attention. These are the conclusions that address the first aim of the article, which was to analyse the prospects for proliferation of shared mobility practices. The second aim was to indicate directions for further practice theoretical developments; the analysis of mobility practices shows that more sophisticated practice theoretical understandings of rules and regulations as well as of breaking of inter-practice links would greatly enrich the practice theoretical toolbox. This article has begun the conversation in this regard. Applying the conceptual tool of ‘suffusing’ to (shared) mobility practices and focusing more on the meanings attached to diverse mobility practices could enrich the conversation further.

Acknowledgements

I thank Ingolfur Blühdorn, Daniel Hausknost, Lisa Ruhrort and two anonymous reviewers for their very insightful and helpful comments on an earlier version of this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The exception is 2020, which showed a decrease in the transport sector’s emissions due to travel restrictions related to the COVID-19 global pandemic.

2 In its beginning, thinking of practices as interconnected entities has been important (Bourdieu 1990; Giddens 1984), and some early theoretical works developed a basic understanding of inter-practice relations (e.g. the concepts of ‘practice bundles and complexes’ in Shove, Pantzar, and Watson 2012 or of practices that learn from adjacent practices in Warde 2005). However, only recently have elaborations became more sophisticated.