Governing smart mobility: policy instrumentation, technological utopianism, and the administrative quest for knowledge

Abstract

This article analyzes administrative practices in relation to the emergence of novel technologies. “Smart mobility” is an umbrella term used to denote the potentially disruptive changes in the transport sector relating to automatization, digitalization, and the platform economy. While this development is largely driven by industry, public administrations are engaging in a number of processes where they seek to obtain knowledge while regulating the development of, for example, autonomous vehicles. The aim of this article was to study how administrative practices of governing create, delimit and constitute smart mobility as a governable object. This is done by analyzing the policy instruments deployed by public administrations to obtain and disseminate knowledge in innovation processes, with the aim of controlling its development. The analysis shows that public administrators utilize four main categories of policy instruments: pilots, standards, scenarios, and collaboration. By developing scenarios, following pilots and collaborating with a variety of stakeholders, the administrations are not only tapping into newly produced knowledge and learning about the potential impact of these technological novelties, these processes are also creating and delimiting smart mobility as an object to be governed.

Keywords:

• Governing
• knowledge
• policy instrumentation
• smart mobility
• technology

Introduction

Western societies are imbued with images of how great technological innovations, such as the steam engine, the printing press and the microchip, have transformed crucial aspects of our lives in fundamental ways (Williams, 1994). Many innovations also develop in tandem with a perceived need of governing. This is expressed first by the wish to understand the innovation and its potential consequences, then by attempts to predict and potentially transform its development in desired directions. In this paper, we explore how public administrators make technological innovations governable, by examining the example of smart technology and platform-based innovations in the transport sector.

Smart technologies largely stem from a number of expansive high-tech giants based in Silicon Valley. Progress has been rapid, and authorities and governments at various levels have often been described as taking a back-seat position, both in relation to access to knowledge about these technologies and in relation to their wider social implications (Mazzucato, 2019; Morozov, 2010). However, public administrations in many jurisdictions are increasingly engaging in, and even initiating, investigative processes with the goal of obtaining knowledge and taking control of the implications of this development. In these processes, administrators tend to face two partly opposing rationalities. On the one hand, there is a rationality where the overarching aim is to protect the public from hazards stemming from unproven technologies. Administrations are then monitoring the developments, to both minimize potential risks and mitigate a future scenario of ungovernability (Offe, 2013). On the other, new technologies and innovation policies are often propelled by a rationality of competitiveness and economic growth. How these rationalities operate and inform policies is an underexplored theme in the literature exploring the intersection of technology and administration. It is at this intersection this paper makes its contribution.

The administration of techno-utopianism—aim and outline of the paper

Despite a renewed interest in the analysis of technology and techno-utopianism in research on administration, there has been relatively little interest in investigating the role of public administration in governing the development of smart technologies (although exceptions can be seen in Martin et al., 2019; Meijer & Rodríguez, 2016; Martin et al., 2019). The point of departure for this paper is that policy formation, including the policies relating to smart technologies, are not perceived as given, but are continuously constructed and negotiated (cf. Manderscheid, Schwanen & Tyfield, 2014). Governing may therefore be understood as a two-sided process that, on the one hand, involves attempts at representing, knowing and delimiting the object to be governed (in this case smart technologies in the transport sector, often labeled smart mobility), while on the other developing and constructing measures and devices intended to transform this object (cf. Paterson & Stripple, 2010). This can be described as a process of creating representations of the “problem” to be addressed, and how it should be understood, including delimitations of what types of knowledge are considered necessary (e.g. technocratic, user based, administrative, etc.). The process may also implicate different kinds of expertise as the definition of the problem is often connected to the epistemic monopoly imposed by certain groups of experts (cf. Bacchi, 2010; Yanow, 1999, 2007, 2015). In this paper, the administrative practices of creating, delimiting and constituting technological innovations into governable objects are studied empirically, by analyzing the policy instruments deployed to govern smart mobility.

The analysis is limited to the introduction of smart technology in the transport sector, a sector currently undergoing potentially large-scale changes in the light of the roll-out of digital technologies. Smart mobility has been used as an umbrella term to describe and promote digitalization and automatization, including automated vehicles (self-driving cars and buses) and platform-based mobility services (Docherty, Marsden & Anable, 2018). To date, the call for more governing in this context is from actors with different agendas: consultants and auto-manufacturers are calling for a review of existing policies and regulatory frameworks, arguing that policies—ranging from traffic regulations to car-parking norms—must be adjusted to facilitate a transition to smart solutions (see e.g. Atkins, 2016; KPMG, 2017). These calls from industry often latch onto national, regional or urban competitiveness agendas.

There are also warnings to policy-makers not to repeat the planning mistakes of earlier decades, where planning was centered around the facilitation of automobility (Urry, 2004). These voices highlight the need to govern rather than merely adapt to technological innovations, thereby emphasizing the importance of assuring that the development of automated vehicles will meet public values such as equal access and reduced emissions (e.g. Docherty et al., 2018; Fagnant & Kockelman, 2015). It is at this intersection between knowledge, public health, technology and competitiveness that smart mobility is currently emerging as a new policy field. By analyzing the policy instruments that public administrations use to understand and potentially control the development of technological innovations in the transport sector, we will be able to explore how smart mobility is being formed as a new policy area and so made into a governable object.

Transport, transition and technology: Situating the study

In the literature on public administration and public policy, there is an emerging interest in understanding how government policy is formed and performed through various instruments (e.g., Le Galès, 2011). However, the few studies published on the governance of smart mobility generally focus on individual innovations, such as autonomous cars (Docherty et al., 2018), autonomous buses (Berglund-Snodgrass & Mukhtar-Landgren 2020), or MaaS¹ (see Finger & Audouin, 2018; Karlsson et al., 2020). As often noted in the scholarly field of mobility studies, transport planning is still, despite decades of critique on techno-optimism, highly “technocentric” (Freudendal-Pedersen & Kesselring, 2016, p. 576). Echoing this, Givoni (2014, p. 1) suggests that policy analysis in transport studies tends to be “concentrated on the impacts of changes in one input variable on travel behavior and the transport system.” A heavily technocratic policy practice is therefore mirrored by an academic discipline based on rationalistic methods such as advanced modeling (Kębłowski & Bassens, 2018). Literature on public administration of technologies relating to innovation within the transport sector is scarce, yet we argue that there are theoretical gains in using the empirical case to theorize policy formation through administrative practices in emerging areas.

In the next section, we outline the theoretical approach, before proceeding to describe the methods and material used. In our mapping, four policy instruments were singled out and are analyzed in depth: pilots and testbeds, scenarios and road maps, collaborative governance, and standards. We conclude by discussing our contribution in relation to how administrative practices are performed through policy instruments, and situate the discussion in broader debates related to governing emerging technologies.

Reconceptualizing policy: From instrument to instrumentation

In the more functionalist literature on public administration, policy instruments such as regulation, taxation or information campaigns are often described as discrete rational tools, or even a “toolkit” that policy-makers utilize to affect behavior and policy outcomes (Bemelmans-Videc, Rist & Vedung, 2011; Grazi & van den Bergh, 2008; Santos, Behrendt, & Teytelboym, 2010). In this paper we follow the argument of Lascoumes & Le Galès (2007), which is based on a critique of more functionalist approaches, where policy instruments can neither be understood as purely instrumental (i.e., driven by a means-ends rationality), nor as separated from the context in which they have been developed. Instead, policy instruments are continuously formed and performed in specific settings, and through their application they create and transform the organization of social relations, as well as the subjects’ understanding of themselves. In other words, “policies both change what we do (with implications for equity and social justice) and what we are (with implications for subjectivity)” (Ball, 2015, p. 306, italics added for emphasis; cf. Lascoumes & Le Galès, 2007, p. 8–9).

Policy instrumentation is defined as “[a] set of problems posed by the choice and use of instruments (techniques, methods of operation, devices) that allow government policy to be made material and operational” (Lascoumes & Le Galès, 2007, p. 4, emphasis added). While the choice of policy instrument reveals, reflects and produces the currently preferred mode of governance, it also reveals something about the context in which that choice is made. The concept of instrumentation captures this contextual and processual approach. In line with our argument above, we understand the processes of “making policy material and operational” as a two-sided processes: on the one hand, a process representing, knowing and delimiting (in this case) smart mobility and, on the other, the measures that are intended to transform this into a governed object (cf. Paterson & Stripple, 2010). We therefore understand policy instrumentation not as a solution to a given problem, but as a processual policy-shaping device worthy of interrogating on its own terms (see e.g. Webb, 2014).

Methods and materials

Delimiting smart mobility

Smart mobility emerged only recently as a concept, and must be understood in light of the recent transformations of the automotive industry. During the financial meltdown 2008/2009, many corporations and suppliers in the automotive industry were on the brink of collapse, or close to filing for restructuring to avoid bankruptcy. Since then, many national governments have tried to find new ways of supporting the development of this industry, not least because the automotive industry, together with its vast networks of supplier chains, provide large numbers of manufacturing jobs. This has prompted novel forms of R&D funding and industry-supportive policies. In these endeavors, a number of different innovation processes are emerging, including electric cars and new solutions for on-demand transport. This empirical material focuses on two of the main smart mobility solutions that have emerged: automated vehicles and mobility-as-a-service (MaaS), as they have developed in Sweden.

The case of Sweden is motivated in two ways. First, its history of combining a strong state and a welfare system with a small but open economy, has meant that it has been heavily dependent on its export industries, including car manufacturing (Larsson, 2002). Recently, both Volvo and Scania have been launching trials and R&D pilots to test autonomous vehicle technology in real-life situations. The government and the national research funders, too, have spread money over environments where smart mobility is researched. The national government’s strategic innovation program, called Drive Sweden, includes a roadmap where automated and connected vehicles are planned for use in real-life traffic by 2022 (Website, Drive Sweden, 2018a). However, issues concerning smart mobility are not only driven by innovation ambitions, but also by sustainability goals (Mukhtar-Landgren & Smith, 2019).

This leads us to the second motivation for the Swedish case. Sweden is often described as a frontrunner in sustainability policies (Lidskog and Elander, 2012). Currently, there is a government discourse on how a transition of the transport sector is vital if we are to meet climate policy goals, where the need to increase the proportion of energy-efficient modes of transport such as walking, cycling, and public transport is often set center stage (SOU, 2013: 84), as a means to improve energy efficiency and make cities more livable in (Swedish Government, 2018).

Mapping and interpreting instrumentation

This paper is situated in the understanding of policy instruments as contextual and continuously constructed (Lascoumes & Le Galès, 2007); accordingly, the empirical analysis is based on an interpretative tradition (Alvesson & Deetz, 2000; cf. Zingale & Piccorelli, 2018). The policy instruments analyzed were related to their context, using previous theoretical and empirical accounts and our own observations. Since this study is part of a larger research project investigating the governance of smart mobility in the context of Sweden and Scandinavia, we already had broad knowledge of both the sector and the area, not least because we have been working on other similar projects prior to this one. Informed by notions of co-production, the occasional input from reference groups and stakeholders has been key throughout the project. The analysis of policy instruments proceeded by way of a systematic categorization of the empirical material in two stages.

First, we mapped the actors and processes involved in development of smart mobility in Sweden. This was delimited to material formulated on the national level within the fields of MaaS and autonomous vehicles, including both self-driving cars and self-driving buses (cf. the definition of smart mobility above). The mapping was conducted between January 2017 and January 2019. The mapping was based on four types of sources and approaches, namely analysis of (i) homepages of key actors and administrations; (ii) gray papers, official documents and investigations from key actors and administrations; and (iii) participatory observations in conferences and workshops with practitioners and stakeholders, such as “meet-ups” organized by innovation agencies, presentations by key players such as Volvo at transport conferences (including the Q&A that followed). While observations are often described on a scale ranging from “participatory” to “observational” (cf. Gustafsson, 2016, p. 49; cf. Fred, 2018), we acted in the capacity of observers, even though we could on occasions speak to participators at meetings, or ask questions in plenum. Besides these sources, we also (iv) conducted a total of 22 qualitative interviews as part of the larger research project, including interviews with staff at the Swedish Transport Administration, several urban planners, one representative from Volvo, Drive Sweden”s director, and other key actors in Stockholm and Gothenburg. Although interviews may serve many purposes (Kvale & Brinkmann, 2009), we used the interviews to point us in the right direction and identify key actors, agencies and policy instruments. In brief, the purpose of the interviews was to obtain background information rather than gather data (Esaiasson et al., 2017). Because of this, the interviews have not been included as primary sources in the empirical analysis of the specific policy instruments. However, this broad mapping process enabled us to identify a malleable administrative landscape, including a diverse set of public actors who were taking on different roles in governing these processes. These are summarized in Table 1.

Table 1. Organizations included in the empirical analysis.

Organization	Role	Webpage	Roles
Drive Sweden	One of Sweden’s 17 Strategic Innovation Programs, commissioned by the government	Drive Sweden.se including newsletter	Administrative platform for funding and promoting smart mobility
Vinnova	Sweden’s Innovation Agency	Vinnova.se	State administration funding R&D and applied research (on innovation)
KOMPIS	Network initiated by the Government's Collaboration Group for The Next Generation’s Travel and Transport (under Drive Sweden, funded by Vinnova)	Kompis.me	Network and platform for promoting collaboration with the area of smart mobility
Government strategy: The next generation's travel and transport	One of Sweden’s 5 “Innovation partnership programs”, commissioned by the government	https://www.regeringen.se/regeringens-politik/regeringens-strategiska-samverkansprogram/nasta-generations-resor-och-transporter/	Government initiated network-building to enable public-private collaboration
Energimyndigheten	Swedish Energy Agency	Energimyndigheten.se	State administration funding R&D and applied research (on energy)
Viable Cities	One of Sweden’s 17 Strategic Innovation Programs, commissioned by the Government.	viablecities.com	Network and platform for promoting collaboration with the area of smart cities. Contributes with “research and innovation in the area of ​​smart sustainable cities” including issues related to (ICT) and digitalization in urban development
Trafikverket	Swedish Transport Administration	trafikverket.se	Following the technical and legal developments internationally
SOU’s	Public official investigations		Adjusting and harmonizing the legislation to enable car-sharing, autonomous cars etc

Secondly, we reviewed the empirical material gathered during the mapping process by searching for policy instruments utilized to govern the development of smart mobility. Inspired by previous research on smart mobility (e.g. Docherty et al., 2018; Paulsson and Sørensen, 2020; Mukhtar-Landgren & Smith, 2019), we used the six policy instruments identified by Lascoumes and Le Galés: (i) legislative and regulatory, (ii) economic and fiscal, (iii) agreement-based and incentive-based, (iv) information-based and communication-based, (v) de facto and de jure standard, (vi) best practices. These were used to organize and categorize the empirical material. We also noted which public agency had used each policy instrument.

These six policy instruments were then re-grouped into four more specific policy instruments: pilots, collaboration, visions, and standards. This re-grouping was the result of an iterative analysis between the empirical material and previous research on governing smart mobility. We noted, for instance, that agreement-based instruments were linked to the ambition of “getting people in the same room” (or project), as all agencies used and emphasized collaboration. Economic and fiscal policy instruments almost exclusively consisted of pilots, as well as best practice. The empirical material showed a high incidence of information-based policy instruments, which primarily consisted of scenarios and road maps, so we renamed this category to better reflect the particular context. The final instrument, standards, also occurred often, but we did not rename this category in the analytical process. The instruments are summarized in Table 2.

Table 2. Results from the survey of policy instruments in the empirical analysis.

Policy instrument	Results from empirical analysis	Key administrations
(i) Legislative and regulatory	Regulations on safety in relation to autonomous vehicles Forwarded/presented international regulations/discussions on regulations Legislation in terms of a potential enabling/obstructing factor is often discussed (e.g. in gray papers, key notes position papers, newsletters). Policy labs related to regulations and frameworks	Transport agency
		Vinnova, Viable cities, Drive Sweden
(ii) Economic and fiscal	Calls for funding opportunities through pilots, projects, programs, testbeds	Drive Sweden, Viable Cities, Swedish Energy Agency, Vinnova
(iii) Agreement-based and incentive-based	Conferences Meet-ups Network-building events	Drive Sweden Viable Cities, Swedish Energy Agency, Vinnova
(iv) Information-based and communication- based	Scenarios Roadmaps Reporting policy developments from other countries or from different companies Newsletters	Swedish Transport Administration Drive Sweden KOMPIS Drive Sweden
(iv) De Facto and De Jure Standards	Reporting on/forwarding international standards	Swedish Transport Administration
(v) Best practices	Encouraging through pilots, projects, test beds Disseminating through newsletters, conferences etc.	Vinnova, Drive Sweden
		Drive Sweden

Processes of policy instrumentation in governing smart mobility in Sweden

In this section, the four policy instruments that emerged during our mapping are analyzed and contextualized. They are presented through an initial brief theoretical contextualization, followed by an analysis of how they are played out in the Swedish context.

Governing through pilots and testbeds

Swedish government agencies are currently investing in different forms of experiments and pilots in the field of innovation and transport. This is in line with international trends of utilizing formats such as Urban Living Labs, testbeds and pilots to facilitate the development of new and innovative services and goods (cf. Evans, 2016; Karvonen & van Heur, 2014). These ventures often have the purpose of developing both “new technologies and new ways of living” (Voytenko et al., 2016) and as policy instruments they are based on both funding/financing (Mukhtar-Landgren et al., 2019), partnerships (Frantzeskaki, Wittmayer & Loorbach, 2014) and on facilitating learning (Warbroek & Hoppe, 2017). Taken together, these ventures have been described in terms of experimental governance with an emphasis on going beyond “business as usual” to test new solutions and generate knowledge (Kronsell & Mukhtar-Landgren, 2018).

Within the emerging policy area of smart mobility, the purpose of pilots is generally to encourage the development of services or technologies, such as autonomous buses or smart mobility services. A large number of pilots are currently under way in Sweden. The innovation agency Vinnova has funded numerous projects, where the last call for MaaS was for pilot projects to “test, evaluate or scale up combined transport services that include public transport services” (Webpage, Vinnova, 2018c). Some of these pilots are carried out in testbeds, delimited urban areas where companies can test innovations and services in what is often described as a real-life environment (cf. Kronsell & Mukhtar-Landgren, 2020).

In addition to Vinnova, the innovation programs Drive Sweden and Viable Cities fund a number of projects, such as Drive Sweden’s project LIMA that “will develop and test a smart, shared mobility service on a selected pilot group of 1000 people working at Lindholmen” (Website, Drive Sweden, 2018b). A related area of innovation is platform technology and handling of data, where there is also a perceived need for more knowledge, not least in relation to security and integrity issues. One example is the project “Data-driven Policy Development” that “develops new methods for collecting and analyzing data, simulating/exploring policy options, using cloud infrastructure and integrating data services and tools […]” (Website, Viable Cities, 2018).

In addition to these technological innovations, policies are also being tested. Policy labs are popular among administrations in several countries within Europe, including the EU Commission’s EU Policy Lab which is “a physical space designed [.] to bring innovation in the European policy-making” (Website, EU Policy Lab, 2018). One example from the field of smart mobility is the Co-creation Lab initiated by the municipality of Gothenburg in relation to Drive me, which worked collaboratively under the theme “How can autonomous transport systems bring value to cities?” (website, Gothenburg City, 2018).

Ambitions may also relate to other goals. This is evident in a policy experiment funded by Drive Sweden, a comprehensive plan for self-driving vehicles. In the press release issued by the municipality, it is stated that “Gothenburg is the first city to develop a plan for autonomous cars” (Webpage, My Newsdesk, 2017), an event that was also marketed on the English home page of the municipality website, and was picked up by international news media. This indicates that pilots are not only launched to generate knowledge and test innovations, but also to market the state or the city in place branding ventures relating to issues concerning competitiveness, growth and benchmarking (Mukhtar-Landgren et al., 2019; von Wirth et al., 2019). Taken together, these pilots and calls indicate the types of knowledge considered relevant for the development of smart mobility, how knowledge should be developed and dissimilated, and which actors are experts in these processes.

Pilots and experiments are a form of policy instrument that involves a number of more tangible tools: public administrators utilize it as a fiscal instrument with tools such as funding, co-funding and provision of land or facilities; and in the process enabling the allocation of resources to certain areas where new knowledge is desired. The accumulation of knowledge also takes place through formal requirements set in calls for funding, as well as through the dissemination of best practices, devices that enable the diffusion and uptake of new knowledge. This is also related to the fact that these innovations are used to market the country as an innovator or even a frontrunner in terms of sustainability or in “smart city” discourse. Finally, almost all pilots are required by funding agencies to involve many stakeholders, as co-production of knowledge is often seen as an intrinsic feature of pilots (Mukhtar-Landgren et al., 2019). This aspect is explored in the next section on governing through collaboration.

Governing through collaboration

Collaboration is an example of how administrators use organization to govern policy development (Sedgwick, 2017; Spekkink & Boons, 2016; Ulibarri & Scott, 2017). This policy instrument is well noted in the literature, where it has also been described as a form of governing in itself, collaborative governance. Collaborative governance has been described as a “deliberate choice to govern” by bringing different organizations together (Vangen et al., 2015, p. 1239). As such, it is an example of governance through “coordinating actors, social groups, and institutions to attain particular goals, discussed and defined collectively in fragmented, uncertain environments” (Hospers, 2013, quoting Borraz & Le Galés, 2010). The tendency to use collaboration as a policy instrument in this way has been described in terms of enabling, facilitating or simply opening up action space for other actors (cf. Mukhtar-Landgren et al., 2019; Qvist, 2017). In the case of smart mobility, it includes enabling various actors (e.g. stakeholders, civil society, industry) to participate in innovation processes (Donahue & Zeckhauser, 2012). The ambition is also to give private actors access to information about, for instance, the regulatory frameworks concerning the development and trialing of, for instance, autonomous buses by opening up and enabling local collaborations.

Collaboration has been described as a key governing tool in the development of smart mobility services (e.g. Karlsson et al., 2020; Mukhtar-Landgren & Smith, 2019). It is used in different ways in the investigated material, from initiation of the large-scale platforms described above in relation to pilots, to smaller formats such as meet-ups. Public agencies, ranging from local to national authorities, frequently host meetings and conferences, and attend events either as invited speakers or as participators. It is also worth noting that several of the organizations and administrations that we mapped are themselves made up of collaborative structures. One example is the 17 government-initiated Strategic Innovation Programs (SIP) (Website, Vinnova, 2018b). One of these, Drive Sweden, has gathered a plethora of stakeholders, including Volvo, Scania and Autoliv, ICT companies like Ericsson, and public authorities. These SIPs, funded by three state agencies, in turn fund and co-fund various pilots within smart mobility, and act as what Lascoumes and Le Galés call “frameworks of agreement, with incentive forms linked to it” (2007, p. 13).

Another example in this context is the government initiative to establish five collaborative platforms in 2015, one of which was the platform Automated and connected vehicles with representatives from universities, industry and government. Public authorities are often members of collaborative platforms. This form of governing by positioning on strategic nodes in networks has been conceptualized in terms of nodality (Hood & Margetts, 2007). A node can be seen as a point in a network, and public actors use its nodality for strategic purposes, for example to influence the transition in a certain direction, or to be kept informed about developments (Hood & Margetts, 2007, Chapter 2). Collaboration is a policy instrument that includes processes of capacity building and pooling of resources, as well as a number of more tangible tools, such as making collaboration a funding requirement for pilots and the organization of events and platforms.

Governing through scenarios and road maps

It’s not all about driverless vehicles. This is a completely new approach to mobility. We are on the threshold of a radical shift, and it’s happening fast. In just a few years the world will change. We will see entirely new mobility business models enabling sustainable cities (Website, Drive Sweden, 2018b).

Scenarios are a policy instrument that occur frequently in transport planning, generally utilizing methodologies such as forecasting, back-casting and Delphi, which are highly engineering-oriented and often based in a technocratic instrumentality (Urry, 2016). In a more general sense, scenarios can have several different audiences and purposes, ranging from being tools to promote urban attractivity to tangible methodologies for problem solving and planning (cf. Mukhtar-Landgren & Smith, 2019).

Like many other countries, Swedish authorities have developed scenarios for smart mobility, in particular for the future of automated and connected vehicles. One was produced by The Integrated Transport Research Lab at the Royal Institute of Technology (Kristoffersson, Pernestål Brenden, & Mattsson, 2017). They organized the development of their four scenarios as a collaborative endeavor involving almost 40 persons from different organizations, including automotive manufacturers, planners, law-makers, and researchers. These scenarios constitute an important part of knowledge building, and the choice of invited actors also tells us something about the expertise and forms of knowledge required.

Besides being involved as a key stakeholder in the development of scenarios described above, The Swedish Transport Administration has collaborated with private and public organizations to increase understanding of “the role of self-driving vehicles in a sustainable transport system and how it affects the Swedish Transport Administration’s planning and forecasts” (Trafikverket, 2018). One such context is DenCity, a scenario-building research project launched by CLOSER² that looked at “innovative types of travel services as an alternative to car ownership” (DenCity Closer, 2018). In addition, Drive Sweden has developed a road map that has been described as an “outlook for personal mobility […] which shows what we want to jointly achieve within our partnership until 2030” (Website, Drive Sweden, 2018a).

Scenarios are a way for public actors to acquire knowledge and signal the way forward to presumptive investors and innovators, but they also help to establish public images about what smart mobility is or should be. When administrators and other experts endorse certain scenarios or roadmaps, the incorporated public images may also delimit future actions and decisions to a limited number of locked-in trajectories.

Governing through standards

The role of public authorities as drafters of legislative and regulatory frameworks is often considered vital for safety and quality. However, the nature of authority is shifting due to increased pressures on public administrations to open up and collaborate (e.g. Paulsson & Isaksson, 2019). In the development of policies on smart mobility, standards are related to a number of aspects, including safety, levels of automation, and levels of readiness for technology adoption. Drawing on observations from a range of studies, Brunsson, Rasche and Seidl (2012) suggest that standards can be recognized on the basis of three characteristics: they are a specific type of rule, they are formally voluntary, and they are intended for common use (see also Thévenot, 2015; Timmermans & Epstein, 2010). When standards regulate processes, they often concern safety or quality processes (Loconto & Demortain, 2017).³ The most well-known standard in this context is the six-level division of automated vehicles.⁴ It originates from the US, but has been disseminated rapidly on a global scale.

This technique is a form of definitional power. The standard was developed by SEA, an international automotive standardization agency, in 2014 (SEA, 2014) and adopted in the US by the National Highway Traffic Safety Administration (National Highway Traffic Safety Administration, 2018). While the six-level division of automated vehicles has created a standard on how to define automated vehicles, it is now working its way into legislation and regulations, thereby reflecting a broader trend where regulations are built upon international standards and the harmonization of rules (van den Ende et al., 2012). Since manufacturers of autonomous vehicles want to lower administrative barriers to innovation, they prefer the same safety rules to be applied across countries, and one approval for piloting in one jurisdiction should be enough for many more (Paulsson & Mukhtar-Landgren, 2019).

In order to facilitate the development of smart mobility, the Swedish government launched investigations in 2016 to assess and evaluate the legal impacts of automated and connected vehicles (SOU, 2016, 2018). As most traffic and transport legislation is based on international conventions and standards, these governmental investigations called for collaboration, or at least close monitoring, of developments at international levels. This would also ensure that the Swedish automotive industry—including manufacturers Volvo and Scania—would not face the risk of lagging behind in terms of competitiveness.

Standards and rules form and preform smart mobility as an object to be governed, by providing formal definitions and frameworks for the development (not least including regulations for testing autonomous vehicles in real-life situations in pilots). As in the case of scenarios, standards contribute to the establishment of conceptual boundaries around smart mobility (e.g. Borraz, 2007), but this also means that the industry-developed standards have worked their way into legislation processes. Such multi-layering is evident in The Swedish Transport Agency’s use of the six levels of automation to assess and evaluate applications for piloting autonomous vehicles (e.g. Lampland & Star, 2009).

Summing up—the emergence of smart mobility as an object to be governed

Policy instruments, taken together, are building the foundation of what is described as an emerging—and radically new—policy area. How, then, is smart mobility being represented and known in these administrative practices? One point of departure for these practices is the uncertainty about the future and the need for more knowledge about smart technologies and their implications. Because many these policy instruments are experimental in nature, not least pilots and testbeds, knowledge is produced when the outcomes of the experiments are evaluated. Experimental governance is often described as necessary against the backdrop of novel and innovative policy-making, where “business as usual” is no longer considered viable (cf. Kronsell & Mukhtar-Landgren, 2018). By adopting standards, administrators are utilizing the definitional power inherent to these, thereby avoiding a situation in which the same concept is used by many actors, but its meaning varies according to context. This is also related to the perceived need for administrations to facilitate the development of smart mobility, taking the step from vision to reality. A number of aspects are perceived to need facilitation here, fostering knowledge production, reviewing regulations, establishing definitional standards, and providing funding, as this is required to test and disseminate the produced knowledge.

A strong emphasis on novelty, uncertainty, and the calls for new innovative forms of policy means that policy-making in this field is not restricted to the formal decision-making bodies. Instead, it is generally unfolding and promoted in informal and temporary collaborative settings (cf. Hopkins & Schwanen, 2018). One observation we made while surveying the material is that a number of key actors during the time we were examining the process were switching from one organization to another, indicating that collaborative governance is a key policy instrument that enables people to change roles, but still continue to meet, discuss with, and befriend each other in workshops and seminars.

Finally, the quest of knowledge is also combined with the need to assert a vision for the future, and one that connotes innovation and future success. To date, almost all organizations involved have developed a road map or a vision of future mobility. These visions might serve different functions—from place branding initiatives relating to competitiveness to signaling priorities to other actors in a complex web of stakeholders. The coherence in visions and models for the future signals a sense of agreement about where we are heading, despite the fact that this development has not happened yet and is largely unknown.

Conclusions

This paper set out to investigate how technological innovations are turned into governable objects and consequently transformed in desired political directions. We examined the devices of governing, in other words, the policy instruments employed to make technological innovations in the transport sector governable. Based on our analysis, and previous research, we suggest that governmental administrations and policy-makers are making smart mobility governable by using four categories of instruments: (i) pilots and testbeds, (ii) collaborative governance, (iii) visions and scenarios, and (iv) standards. Together, these policy instruments aim to delimit, represent and understand the processes at hand, while involving measures and devices to transform this governed object, a process still very much in its infancy. We describe this as a process where policies—in this case smart mobility— are being formed and performed.

These results speak to broader debates on theory in public administration and public policy in three ways. First, the policy instruments we have identified are building the foundation of what we have described as an emerging—and radically new—policy area, that of smart mobility. Even though the interest in technology and techno-utopianism in research on public administration has grown during recent years, relatively little interest has been devoted to investigating how new policy areas are emerging and how governmental policies are formed and performed through the development and administration of policy instruments.

Secondly, since the implications and potential failures of technological innovations in the transport sector may cause harm and undermine traffic safety, these have prompted public administrations in Sweden (and elsewhere) to develop and employ policy instruments that either produce, or gather existing, knowledge. Policy instruments such as pilots and testbeds are clear examples of how experimental forms of governance are used to generate knowledge and enable learning (e.g. Zingale & Piccorelli, 2018). When “business as usual” is no longer considered viable, in this case in relation to a future of smarter mobilities, experimental governance is often perceived to be necessary to guide policy-making.

Thirdly, policy instruments were deployed in ways that were not only intended to produce knowledge about the technological innovations, but also to transform smart mobility futures. The policy instruments used indicate a process of governing smart mobility so that it facilitates national, regional or urban competitiveness, while keeping an eye on how these technological innovations potentially impact traffic regulations and international standards on automotive safety. In these experimental and collaborative forms of governance, formal policies and potential contentions were not present to any great extent, and the policy area was largely framed in technical terms, which was mostly visible when it came to safety issues.

The policy instruments used propelled an already widespread techno-utopianism surrounding the development of both autonomous vehicles and mobility-as-a-service. At the same time, there was an awareness among policy-makers and administrations of the limitations and potential failures of these technological innovations. As shown by the choice of policy instruments, the risks and hazards coupled to the roll-out of autonomous vehicles were taken seriously. However, Morozov (2010) and Callen and Austin (2016) have pointed out that policy-makers often believe that they can overcome potential limits and failures by simply appearing as serious and by acknowledging the shortcoming of a particular technology. Ambitions to govern smart mobility can therefore be understood as an expression of a double-edged desire: to protect the public from hazards and risks stemming from new and unproven technologies on the one hand, and to allow unproven technologies to be tested and piloted, and eventually developed and potentially contributing to growth on the other. After all, unless new and unproven technologies are transformed into, and treated as, governable objects, there is a perceived risk that new technologies will develop autonomously and possibly make society, or at least parts of it, ungovernable (Offe, 2013).

In our case, the administrators’ use of experimental and collaborative policy instruments may be understood as a will to obtain knowledge about possible shortcomings, thereby ensuring that smart mobility futures could be governed in certain desired directions (e.g. for enhancing national, regional or urban competitiveness). This opens up broader questions about the knowledge/power nexus in (public sector) administration of innovation and technology, and what kind of knowledge is to be understood as proper and useful knowledge. But these questions will be left for future studies of policy instrumentation to explore.

Additional information

Funding

This work was supported by Energimyndigheten.

Notes on contributors

Dalia Mukhtar-Landgren

Dalia Mukhtar-Landgren, PhD, is a researcher and senior lecturer at the Department of political science at Lund University in Sweden. Her research includes new forms of urban planning and development, including issues such as projectification, experimentation and innovation work. She is currently engaged in research projects on testbed planning, urban experimentation, smart mobility and processes of local development planning.

Alexander Paulsson

Alexander Paulsson, PhD, is a researcher and senior lecturer at the School of Economics and Management at Lund University, Sweden. He is currently doing critical research on the policy and governance of smart mobility as well as on the marketization of public transport. His research interest span organization studies, social studies of science and technology, and ecological economics.

Notes

1 MaaS, or Mobility-As-A-Service is an integrated mobility service that offers the opportunity to combine different transport modes into a one-stop-shop in an app, with the overall ambition to create a seamless mobility experience (Karlsson et al., 2020).

2 CLOSER is the Swedish national arena for collaboration within transport efficiency at Lindholmen Science Park in Gothenburg

3 In the transport sector, national governments have traditionally been preoccupied with issuing permits and licenses to corporations and individuals as a way to authorize them and enable them to use certain forms of technologies (e.g. different forms of vehicles) and infrastructures (e.g. road, rail, waterways and airways). As safety is key in this sector, any violations of national and international safety regulations may mean that permits and licenses are withdrawn, either permanently or temporarily. Quality is assured though these permits and licenses, while actual outcomes are difficult to assess, though law enforcement bodies may occasionally check whether vehicles and drivers comply with the legislations.

4 The formal and standardized definition of the six levels of automation currently used in Sweden as in many other countries is (0) No Automation, (1) Driver Assistance (2) Partial Automation (3) Conditional Automation (4) High Automation (5) Full automation.