Decreasing the share of travel by car. Strategies for implementing ‘push’ or ‘pull’ measures in a traditionally car-centric transport and land use planning

Abstract

This paper analyzes strategies that can be successfully pursued to implement measures to reduce car traffic in what has traditionally been a very car-centric planning praxis. Analytically, the paper use path dependency theory to provide an understanding of why certain types of measure are not implemented in cities on as widespread a basis as policy objectives may require, and to understand how transport planning path dependence in urban authorities might be changed. Empirically, the analysis builds on a comparative case study of transport and land use planning in Swedish cities. The most effective strategies do not appear to be radical policies leading to fast implementation of goals about sustainable transport, for example by implementing very car restrictive measures, even in the face of resistance from the public and from within the city administration. The results support an approach that from a strategy making perspective can be understood as an institutionalizing process by which internal organizational and external public support for car restrictive and potentially controversial measures are built. Implementation may be achieved by building new institutions within city administrations, where routines and norms gradually change so that car restraint measures gradually become part of the normal way of doing transport planning. This then starts to lock-in certain patterns of travel and make further car restraint measures more feasible and institutionalized as part of a standard menu of measures that cities use, and not something out of the order.

Keywords:

• Traffic restraint
• car restriction
• transport planning
• policy
• sustainability
• path-dependence

1. Introduction

Transport and land use planning have since World War Two been planned with a focus on the private car. Admittedly, politicians in many European cities have the last decades adopted goals intended to decrease the share of travel by car in relative or absolute terms in order to reduce energy use, emissions, accidents, and noise etc (see Rye & Hrelja, 2020 for an overview of policy goals and proposed measures in a sample of West European cities). However, research shows how difficult it is to change the direction of planning and stop planning for the car and start planning for walking, cycling and public transport. In cities, the sustainable transformation of transport systems is often constrained by barriers of rebound effects, conflicting visions at different levels and lack of consensus among stakeholders leading to continued planning for cars (Nikulina et al., 2019).

This paper addresses a central issue in the transition toward sustainable transport and land use planning. How can cities manage and implement goals of less car use? The paper resonates with recent research that analyses how to plan for decreases in car use, in what has traditionally been a very car-centric planning praxis. Research has shown what measures that have typically been implemented by those cities that have reduced car use as a proportion of trips (see for example Buehler & Pucher, 2011; Buehler et al., 2017a; Buehler et al., 2017b for European examples). However, the important question about implementation strategies and how to initiate or accelerate a transition toward a planning that favors traffic restraint, walking, cycling and public transport still need more analysis, as there is no general agreement on the best ways to pursue this change (Nikulina et al., 2019). This is partly a consequence of the traditional focus on the means and tools of policy making in transport research. Transport (policy) research lacks, as Marsden and Reardon (2017) point out, analysis about the actual realities on the ground, such as context, power, resources and legitimacy, that is, factors that influence transport and land use planning practices.

There is thus a need for a more reflective discussion on the ways to decrease the share of car traffic. The purpose of this paper is to contribute knowledge of those strategies that can be successfully pursued to implement measures to reduce car traffic. Specifically, the paper seeks to find out:

• What measures cities need to implement to achieve the goal of reduced car use.

• Why these measures are often left out of plans completely.

• Why these measures are often not implemented even if they are included in plans.

• What can be done to increase the probability of them being implemented.

Analytically, the paper use path dependency theory to provide an understanding of why certain types of measure are not implemented in cities on as widespread a basis as policy objectives require, and to understand how transport planning path dependence in local authorities might be changed, so that the organization gradually shifts to a different, less car-dependent focus in planning. Empirically, the analysis builds on a comparative case study of transport and land use planning in two Swedish cities – one assessed by peers in the transport planning sector to be ‘leading’ (Lund) in its approach to reducing car use, and one more ‘average’ (Eskilstuna). Both have clear and time limited objectives to reduce the proportion trips made by car, but they exhibit great differences in how successful they have been in achieving the goals (a much deeper explanation of the reasons for selecting these cities is given in Section 3). The cases illustrate politicians’ and planners’ reflections on how the political goal of reduced car traffic can best be achieved, and how potential barriers and enablers influence the choice of strategies and measures.

2. Literature review of planning practices and measures

There are several different types of measures that can be used to reduce the proportion of travel by car (see Marshall & Banister, 2000 for an overview). A distinction can be made between measures that push passengers to other modes of transport, by making these more attractive and the car a less attractive option, and measures intended to pull passengers to walking, cycling and public transport (Marshall, 1999). For example, integrated transport and land use planning may decrease the need to travel by car and make it easier to walk and bicycle if distances are short (Naess et al., 2013; Van Wee & Handy, 2016). Another approach is to use different forms of financial and behavior-changing measures such as taxes, and fees, or information, marketing, and behavior campaigns – so-called mobility management (Nash & Whitelegg, 2016). Yet another way is to decrease the physical space for cars in the city, for example, through restricting speed, reducing road capacity, and reducing the number of parking spaces. Several researchers argue that measures to decrease the share of travel by car must be implemented in parallel with measures to increase travel by public transport, walking or cycling if modal share is to be changed to be less car-dependent (Owens, 1995; Marshall & Banister, 2000; Banister, 2008; Nash & Whitelegg, 2016).

Some of the earlier research contains, as explained earlier, empirical descriptions of measures for reduced car traffic that have been implemented in European cities and how these have achieved their objectives (Buehler & Pucher, 2011; Buehler et al., 2017; Buehler et al., 2017, Granås Bardal et al., 2020). This research show that the success of cities as Munich, Hamburg, Vienna and Zurich in reducing the car share partly can be explained by integrated policy packages where the implementation of the above-mentioned measures produces synergetic impacts (Buehler et al., 2017). However, there are also numerous studies in many cities that describe how planning decisions concerning their physical development are taken in a way that undermines the ambition to deliver a sustainable transport system, with the result that local transport planning continues along the same traditional car-centred trajectory. These previous studies show how measures for reduced car traffic, and indeed in relation to that measures that somehow restrict car traffic (road space reallocation, for example) are potentially contentious and therefore often experience implementation problems (Hrelja et al., 2013; Isaksson & Richardson, 2009; Nash & Whitelegg, 2016; Sørensen et al., 2014). This may be the reason why, for example, Swedish cities are primarily seeking to reduce the share of car traffic by making walking, cycling and public transport more attractive relative to car use. Very few Swedish cities are trying to reduce car use by means of direct restrictions, reduced road capacity or reduced accessibility for car traffic (Hrelja, 2019). This is also the case in several Dutch, German and British cities, where the only means proposed of cutting car use is on-street parking management (Rye & Hrelja, 2020).

The research describing implementation problems also indicates the need to go beyond an empirical analysis of what measures to reduce car traffic have been implemented in cities, and better understand how the ‘actual realities on the ground’ influence decisions about measures, especially the choice between push or pull measures or ways to implement potentially controversial car restricting measures. Previous research illustrates how transport planning practices are performed by routinized practices embedded in power relations and discourses that hold them in place despite attempts to change them (see for example Richardson, 2001, Low et al., 2003). The concept of path dependence can be used to better understand how such factors produce ‘lock-in’ pathways of planning decisions. The concept of path-dependency has been used in several scientific disciplines (see for example Sorensen, 2015 for an example from the planning research field, and Poku-Boansi, 2020 for an overview of the concept in a transport research context). This paper uses a definition that draws inspiration from previous transport research. The difficulties in implementing radical changes in transport planning practice (among which changes car restraint measures such as road closures, parking management and so on can be enumerated) have been explained with reference to three interrelated factors that create path dependencies in transport planning and decision making (Low et al., 2005; Hensley et al., 2014). These are:

• institutional factors, relating to practices, routines, and methods applied by key organizations, and the relationships between the key actors in those organizations;

• technical factors, relating to the momentum resulting from fixed infrastructure serving societal functions – for example, a transport system that is very road-based makes it very difficult to provide other forms of access, and

• discursive factors, relating to assumptions, justifications or beliefs, that apply within an organization and shape its practices. An example of this might be the often unchallenged assumption that building new transport infrastructure leads to economic development.

Institutional and discursive factors are strongly related, but can be distinguished: the first refers more to standardized ways of operating, both within and between organizations; and the second refers more to the ‘storylines’ or discourses that develop to either consciously or unconsciously justify those standardized ways of operating (Hensley et al., 2014). In this paper we hypothesize that institutions and discourses in many urban (transport) local authorities establish cultures of practice and predictable patterns of actions, plus discourses, that lead to the continuation of a focus on planning for cars; and both need to change if that focus is to shift to more sustainable transport alternatives.

2.1. Strategies for breaking path dependencies

There is a range of different strategies that can be used to break away from this path dependence that continues the focus in planning practice on planning for cars (Buehler et al., 2017, Bratzel, 1999, Hysing, 2009, Pflieger et al., 2009). Previous research offers some knowledge about how to initiate change in planning practices, such as reorganization of local administrations and the important role of green policy entrepreneurs in local authorities (Hysing, 2009). Bratzel’s (1999) study of cities in Switzerland, the Netherlands and Germany that have been ‘relatively successful’ in greening their transport policies identifies a number of factors that are important in bringing about change, including the seriousness of the problem to be solved; institutional structures and their capability to change; socio-cultural framework conditions, such as public awareness of environmental issues; the different actors involved and their relative (political) strength; and sudden unexpected events, such as environmental or road safety crises. Bratzel’s (1999) paper, and that by Pflieger et al. (2009) that features case studies of transport policy change or inertia in six French, German and Swiss cities, both emphasize the importance of major events, such as the arrival of new actors, or major changes in political direction, as key initiators of change in transport policy direction, a point that will be returned to later in this paper in relation to its Swedish case studies (see section 5.1. where the analysis of Lund and Eskilstuna is discussed in relation to previous research about other cities).

Related to this, an important question addressed by this research is which strategies should be used to implement transport measures that are contentious because by their nature they seek change in travel behavior and therefore challenge traditional norms, routines, and interests in urban authorities.

A theoretical question of relevance for the understanding of the ways to decrease the share of travel by car is, a) should the strategy be to focus on creating internal organizational and external public support for car restrictive and potentially controversial measures, or b) should politicians and planners work in a more narrowly focused way to engage with key actors to develop a more radical policy leading to faster implementation of goals about sustainable transport, for example by implementing car restrictive measures, maybe even in the face of resistance from the public and from within the city administration (Banister, 2003; Isaksson & Richardson, 2009; Fenton, 2016)?

The first strategy implies a way of working where urban authorities would initiate deliberative processes among politicians and officers, not only concerning a policy shift toward more sustainable transport modes, but also to gradually reduce conflicts in interests, norms, habits, and attitudes to reducing car traffic between politicians and officers (Hrelja, 2015), which would perhaps take a long time and risk being watered down through negotiations, possible consensus-building procedures or compromises. On the other hand, more disruptive (Thaller et al., 2021) or radical approaches to sustainable transport require strong support, over time, from political leaders, who will often need to put a lot at stake and take risks in defending controversial approaches (Attard & Ison, 2010; Isaksson & Richardson, 2009). Additionally, research shows that radical approaches may fail if there is a lack of sensitivity to and absence of strategies for working with the more subtle and deeply rooted power relations connected with norms, habits and attitudes that are often strongly institutionalized within cities’ planning and transport departments (Hrelja et al., 2013).

In summary to this section, then, there is a strong analytical framework based on the following factors:

• Path dependency as a mechanism that limits or slows the adoption of more radical measures to restrain and limit car traffic – can it be observed, and what are the mechanisms that produce it?

• Changing the path of transport planning in cities – how can this be done, and how much does it depend on directly addressing the mechanisms that are observed to cause path dependency?

The case studies in this paper have been selected deliberately so as to be able to cast some light on these theoretical questions. In the empirical part of the paper, we describe the measures that have and have not been implemented in these cities and explain in a political sense how their implementation was taken forward in the face of traditional car-focused infrastructure, institutions and discourses. In the final synthesizing section, using these cases, the paper discusses the relevance of its theoretical questions to the overarching issue of how more car-restraint based policies can be adopted and implemented, and draws policy recommendations from this. First though, in the following section, an account is given of methods and sources.

3. Material and methods

The results are based on case study methodology. By using a case study method, the paper provides empirically grounded descriptions sensitive to contextual conditions about strategies for decreasing the share of travel by car. The importance of actors and social relations in strategy making in urban authorities calls for case studies, because with the case study method one can study planning in its context (George & Bennett, 2005; Yin, 2009). The case study method is focused on explaining or understanding a case in the sense of a functioning complex whole, which is studied in its context.

The results are based, as has already been described, on an analysis of two cases (Lund and Eskilstuna). A study of two cases cannot be used to making statistical generalizations. Instead, the general conclusions that are drawn build on the concept of analytical generalizations (Yin, 2009), which means that the findings from the two cases are related to findings in existing research from the field. The analysis of the cases provides empirically based examples of planning, with the ambition of using the cases to understand and develop a theoretical understanding of strategies for decreasing the share of travel by car. The case study descriptions of Lund and Eskilstuna also offer officers and politicians real world planning examples that may inform planning praxis. As Flyvbjerg points out, common to all experts is that they operate on the basis of intimate knowledge of several thousand concrete cases in their areas of expertise (Flyvbjerg, 2006, p. 222). Through the chosen case study method, and the choice of cases, the paper can hopefully explore emerging practices and identify innovative practices in urban authorities of relevance for the development of planning expertise (Fischler, 2000, p. 194). One of the case study city, i.e. Lund, can be considered as a leading urban authority, at least in a Swedish context (see selection of cases below).

In this paper an information-oriented selection of cases has been carried out. This strategy for case selection is used with the ambition to maximize the utility of information from small samples or even single cases (Flyvbjerg, 2006). Lund and Eskilstuna belong to a population of urban authorities that consists of 21 so called ’larger Swedish cities’. Larger cities are defined in Sweden as municipalities with at least 40,000 but fewer than 200,000 inhabitants in their largest urban area (SALAR (Swedish Association of Local Authorities & Regions), 2017). These cities are large enough that a customer base does exist for public transport that can offer a real alternative to the car as a mode of transport, with the result that they are of interest from an analytical standpoint compared to smaller Swedish cities.

The sampling from the population of ’larger Swedish cities’ was based on pre-selected criteria, listed below.

• Trends in travel patterns (both cities where trends in travel patterns have become more sustainable; and those that have followed the national pattern).

• Political goals that explicitly deal with the distribution of transport modes including car traffic.

• Cities’ work with specific areas of action, such as parking or prioritization of transport modes.

• The ambition was also to include urban authorities that can be considered to be either ‘leading’ and ‘average’ cities – (in the sense of being seen in by peers in Sweden as leading or average within the field of sustainable transport).

First, all 21 cities’ political goals and planned measures to decrease the share of travel by car were mapped. The material analyzed were comprehensive plans, traffic strategies/SUMPs (Sustainable Urban Mobility Plans), parking programs, action programs etc. Cities that had clear, or quantitative and timed goals for reduced car traffic were given priority. This was followed by a review of the cities’ surveys of travel patterns (RVU) with a focus on trends in the modal split over time. If no RVU was available, the city was removed from the selection list. Cities with a decreasing share of travel by car or low shares of travel by car were prioritized, but also cities that followed national trends, or had an increasing proportion of travels by car despite the fact that there were political goals for reduced car traffic, were identified. Many cities were excluded in this stage because they did not excel either in terms of goals or travel trends. Finally, three officers from the three cities of Örebro, Västerås, Norrköping (who participated in a reference group associated with the project which this paper is written within) were consulted, and were asked to make suggestions on cities that can be considered as either ‘leading’ or ‘average’ in Sweden. This selection process resulted in the choice of Lund and Eskilstuna (see Tables 1 and 2). The commentary below shows the similarities and differences between the two cities.

Table 1. Transport policy goals and proportion of travel by car.

City	RVU	Development of travel	Goal for reduced car traffic	Leading/Average
Eskilstuna	Eskilstuna Municipality (2016)	High proportion of travel by car (64%). Increasing proportion of car journeys 2010–2015 (Eskilstuna Municipality, 2016).	The goal is to reduce car traffic to 39% by 2020 (Eskilstuna Municipality, 2016)	‘Average’
Lund	Lund Municipality (2018a)	Low proportion of trips by car, 34%, (Lund Municipality, 2018a). Within central Lund 25% (Lund Municipality, 2014).	All travel increase in Lund municipality must take place by foot, bicycle or public transport (Lund Municipality, 2018b). Of all travel within Lund Municipality, 70 % is intended to be by public transport, walking or cycling by 2020, and 75 % by 2030 (Lund Municipality, 2014).	‘Leading’

Table 2. Selected transport characteristics of cities (Public Transport: PT).

City	Population of municipality (year of data)	Mode share % all trips by:				Car Ownership Per 1000 people (2019)
		Car	PT	Bike	Walk.
Eskilstuna	105,000 (2020)	64	11	13	11	461
Lund	120,000 (2020)	34	23	27	14	522

Note: For data sources see Appendix A.

3.1. Introducing the municipalities, their transport policy goals and measures

Lund (population 110,000 approx.) occupies a central location in southern Sweden, some 20 km from Malmö (Sweden’s) third largest city) along a vital regional communication route with large commuter flows. Transport policy goals in Lund are based on the decision taken by the City Council to aim to halve greenhouse gas emissions by 2020 and to achieve an emission level close to zero by 2050. Motor vehicle traffic per inhabitant is to decrease on an annual basis. CO₂ emissions per person generated by traffic are to decrease by 2.5% per year (Lund Municipality, 2014). The future city imagined is a city where:

Walking, cycling and using public transport constitute the main part of all transport activity […]. The need trade and industry have for accessibility has been met in a sustainable way (Lund Municipality, 2014, p. 11).

The above quotation illustrates that there is in Lund a desire to change the approach taken to traffic and city planning by no longer planning for (car) mobility but instead planning for accessibility (Lund Municipality, 2014, p. 7), roughly understood as the potential for reaching work places, community services and trade with ease or within a given travel time. The second is to use measures that make walking, cycling, and public transport more attractive relative to car use. For walking and cycling, it is about creating a high standard and attractiveness in the transport system as a whole. Investments in road infrastructure may be necessary, but investments should primarily go to sustainable modes of transport. Investments in high-speed cycling and cycling streets are important according to the city, and bus traffic should be developed with more frequent departures and given higher priority in the traffic environment with its own lanes and priority in crossings according to the idea ‘think tram – run a bus’ (Lund Municipality, 2018b, p. 41). Finally, a reduction in car travel also presupposes integrated land use and transport planning (described by the municipality as urban planning) that result in densification and a mix of functions in locations with good conditions for public transport (Lund Municipality, 2014, p. 13, Lund Municipality, 2018b, p. 37). The ambition is that urban areas shall primarily grow where there are strong public transport and bicycle traffic routes to reduce transport needs and improve the conditions for sustainable travel (Lund Municipality, 2018b, p. 41). It is worth noting that there is not any particular commitment to greater restraint of car traffic through road closures, more parking management or road pricing.

Also in Eskilstuna (population 105,000 approx.), situated in southeast Sweden, there are clear transport policy goals aiming at a reduction in car traffic. The goal is to reduce car traffic to 39% (Eskilstuna Municipality, 2012, p. 6). The ambition was initially to increase the proportion of trips by bus and bicycle from 8% and 13% to 16% and 26%, respectively, by 2020. The measures proposed and used have not been able to curb the increase in car travel. The share of car traffic has increased since 2010, and the municipality has changed the year when it wanted to achieve the goals of reduced car traffic from 2020 to 2030. The goals are to be achieved (according to the comprehensive plan):

By planning for reduced need of using cars, the car’s share shall decrease in favor of walking, cycling and public transport. Car use generated by future land use developments must be lower than today's average for car journeys per household (Eskilstuna Municipality, 2013, part 2, p. 48).

The measures proposed to be used intends, in summary, to reduce the need to use cars (Eskilstuna Municipality, 2012, p. 50) by making public transport, walking and cycling easier. The city of Eskilstuna wants to reduce the need to use car by planning for mixed land uses, planning residential buildings and offices in connection with important public transport nodes and routes (Eskilstuna Municipality, 2012, 2013), and by giving priority to walking, cycling, and public transport wherever they compete with car traffic for space (Eskilstuna Municipality, 2012, p. 48). The city also wants to improve the supply of public transport to increase attractiveness and travel (Eskilstuna Municipality, 2012, 2013).

3.2. Interviews and written material

To analyze the strategies for the implementation of the goals in Lund and Eskilstuna, we will use qualitative material that consists of both interviews and written material. The written material comprises comprehensive plans, transport, and parking strategies (Eskilstuna Municipality, 2012, 2013, 2016; Lund Municipality, 2018b, 2014). The plans were used to identify transport objectives and the measures selected to try to achieve the objectives (see section 3.1).

The interviews were semi-structured, based on an interview guide. The aim of the interviews was to elicit the subjects’ views on the achievement of each city’s transport objectives, and what choices have been taken in each city in working toward those objectives. Qualitative interview studies normally have only a small number of subjects, but on the other hand they attempt to examine the research topic more deeply and within its context. Eight officers and two politicians were interviewed in 2019. Additionally, the case study results for Lund are partly based on additional interviews (three politicians and five officers) from a previous research project published in Hrelja (2015). These interviews are of relevance for the research questions because their content shows how the interviewees perceive the pursuit of transport policy objectives to take place in Lund, and how they view working methods and relationships between different parts of the administration that are of relevance for transport policy making and implementation.

The interviews were recorded, with the approval of the interviewees, and transcribed verbatim. Each interview lasted for roughly 1–1.5 hours. Interviewees were primarily selected based on their roles in municipal administration and politics. The ambition was also to interview people from different political parties and departments. All interviewees have a designated responsibility for transport or land use planning in their organization. In Lund politicians from parties both in the majority and in opposition were interviewed. The politicians include, for example, the chairperson of the City Planning Committee and a Deputy Opposition Leader. In Eskilstuna, representatives of two different parties in the political majority were interviewed. The politicians include the chairperson of the City Planning Committee and an executive Member for City Planning. The officials interviewed work as heads of transport or land use planning departments, as town planners and traffic planners at the City Planning Office or handle transport planning in the Technical Service Office.

The questions asked touch mainly on the following fields:

• How have Lund and Eskilstuna worked over the past ten years to make travel in the city less car dependent?

• What strategies and measures have been used, and what measures have been implemented, to reduce car travel?

• Are there factors that make it difficult to implement measures to reduce car traffic or that conversely make it easier to do so?

• In view of your answer to the previous question, what is the ‘best way’ to implement measures to reduce car traffic? What conditions must exist, and what methods and working methods should be chosen?

The focus in the analytical work is directed toward understanding how interviewees see the conditions for implementing goals of reduced car trips in the urban authorities they work and, based on this, discuss how their actions can be understood from a strategy making perspective. The research and analysis process has been inductive in the sense that on the basis of the interviews it tries to arrive at descriptions of different ways of implementing goals for reduced car traffic that have emerged in two urban contexts with the different conditions that exist in the cities and that affect the strategies that have emerged.

The analysis was made stepwise. First, all interview transcripts were read through city by city, and parts perceived as containing important reasoning and descriptions from the interviewees was marked. Then all the transcripts were read through several times in depth. Themes, here understood as recurrent regularities in the material (Ryan & Russel Bernard, 2003), were identified in this way. Differences, i.e. the interviewees’ different interpretations and experiences, were also identified. In the result section, quotes from the interviews are used to illustrate the analysis.

4. Results

4.1. Problems moving away from car-focused planning praxis in Eskilstuna

From the description of the transportation context of Lund and Eskilstuna above it can be concluded that there are some similarities between the cities regarding what measures they say they want to implement to achieve transport policy goals. One similarity is that car mobility is not really put at stake in the design of policy interventions, as it seems. In this they are not different from most other Swedish cities (see Hrelja, 2019). Despite the lack of (current) measures to restrict car use, Lund has been very successful in decreasing the share of travel by car (compared with other Swedish municipalities). However, the empirical presentation below will begin in Eskilstuna, where the reason for the increase in car travel in Eskilstuna, despite the political goals to the contrary, according to several interviewees, is precisely the lack of car restrictive measures. In the empirical description that follows, we shall first describe how transport planning practice in Eskilstuna has found it difficult to leave behind its traditional focus on car use. We will then outline how Lund in contrast has gradually developed new ways of understanding how it can achieve its objectives.

According to the officers interviewed, transport policy and planning in Eskilstuna is still characterized by a conventional transport policy and planning approach lacking car restraint measures. Although the City’s transport policy goals indicate political ambition to change direction, planning in practice continues to follow a ‘car-friendly’ path because of politicians’ decisions. An officer in a managerial land use planning position said that there is agreement on the goals, but that:

There are differences between officers and politicians on how to achieve the goals and which measures steer in the right direction or in the wrong direction… From the politicians’ point of view, they think that, if we favour the car a little does not mean we are driving in the wrong direction, while some officers think that; well every time we favour the car we take a small step in the wrong direction. We see in the statistics [on the development of transport mode share] that we are not going in the right direction. We need to do something. We need to do something more drastic if we are to achieve the goals.

The quote illustrates how the officers interviewed perceive there to be several factors hindering the implementation of car restraint measures, a key one being politicians’ continued support of car traffic. According to these officers, there is opposition to car restraint measures from certain ‘loud’ residents to whom politicians are considered to listen too much, but also unsympathetic or even active opposition from some officers within the municipal administration.

The measures for decreasing the share of travel by car that are implemented are considered too few and too weak to break the increase in travel by car, below illustrated by an interview with a transport planner.

The city's largest workplace, Mälar Hospital […] does not have separate cycle paths to its entrance. It is mad that we in 2019 have not succeeded in giving employees the conditions to cycle safely and with good accessibility to their workplace. It is structurally completely reprehensible. What is required is measures […] where the car is allowed to stand back, such as reduced lane width […] and bus lanes […] These are quite big steps that need to be taken. We can take many small steps in individual projects. But they are too small and too few and do not affect so many. They are very good, but they are ant steps.

Officers working to implement measures to reduce car traffic thus feel that more measures are needed to implement the political goals at a faster pace than is currently the case. When it comes to actual implementation, ‘we back off’ as one land use planner put it. The officers interviewed try to handle situations when measures decreasing the share of travel by car are questioned by ‘becoming even better and more pedagogical in presenting evidence-based documentation in the right way to the right people’, as a transport planner put it. The ‘strategy’ they use to overcome skepticism and to speed up the implementation of measures in individual planning projects is to be methodical, have good arguments and to produce planning documents that justify the implementation of measures in an evidence-based way.

A concrete example that several officers use to illustrate this strategy, and the failure of it, is the planning of Munktellstaden. Munktellstaden is a district in central Eskilstuna, which was originally an industrial area that in recent years has been transformed into a district with housing, museums, restaurants, a swimming pool and the multi-sports arena Munktellarenan. When planning the arena, the officers tried to show the politicians that additional car traffic could be managed without having to build a parking garage by using mobility management measures, improving bicycle parking and providing better public transport etc. A town planner described that there:

[…] was a long planning process where we [officers] showed that the parking could be solved in a different way [than through a parking garage]. One reason why we put an arena in the center of the city was precisely because it could be accessible by sustainable transport modes. We ran it all the way and then when everything where ready, the politicians decided to build a parking garage. It feels a little sad, of course, because I think it was very much based on anxiety rather than an actual need. This is something that officers must accept […] In Eskilstuna, politics is not very brave. It’s business as usual. I feel very frustrated. I’m a bit like [the Swedish environmental activist] Greta Thunberg. No, we do not have time for this, we do not have time for this!

Officers see the planning of the Munktellarena as an illustration of the need for more dialogue between politicians and officers in order to create consensus on goals and on the implementation of measures for decreasing the share of travel by car. However, politicians have a completely different view of the conditions for the implementation of measures as well as the pace of implementation, and also perceive the officers to lack an understanding of what is politically possible to implement. The planning of Munktellstaden is also an example of this, but from the politicians’ point of view, below expressed by the chairperson of the City Planning Committee, reasoning about the planning of Munktellstaden from a user perspective:

If I am a mother with two children, and we are going to go swimming with all our swimming gear, then I want to ask which officer wants to walk a kilometer […] with his children to go and swim. Our visitor numbers at the bathhouse show that it is not well attended because visitors do not get there. We also tried handball event buses when [the city’s handball team] is playing and the arena is full. No, people do not take the bus, they take the car on a weekday evening. There was nowhere to park and there was chaos on all streets, which irritated all Eskilstuna residents. When they came home from work, they did not get a place to park because it was a match in the arena. The officers have a goal of a car-free area but they do not see the finances for the bathhouse, and they do not see the activities for the children and how the children shall get to different trainings […] You could say; no, a parking garage should not be built, the area should be car-free! But then we would not have any visitors, and then we would not have had to build the bathhouse and not the arena either.

For this politician, planning must be based on a balancing of different interests and not only the goals of decreasing the share of travel by car. In a similar way, an executive Member in the City Planning Committee expresses that as a politician you must be ‘pragmatic’ when it comes to the choice of measures:

I think you must have carrots, you cannot only use sticks. You cannot just say that we should make motoring more difficult. If you do not offer a good alternative then the motorists endure very, very long before they change.

Although this politician does not believe that the city will achieve its transport policy goals to 2030, at the pace and through the measures that are now being discussed and implemented, he still believes that more radical measures against cars should be avoided, measures that would be politically difficult to implement. The officers’ ‘strategy’ for overcoming skepticism and speeding up the implementation of car restrictive measures in individual planning projects, by being methodical, well-argued and producing evidence-based planning documents that justify the implementation of measures, will therefore probably not succeed. It is not the lack of knowledge that is the ‘problem’.

This description of planning in Eskilstuna raises a number of analytically interesting questions about how to establish well-functioning working relations between politicians and officers, as this appears very important if measures to decrease the share of car travel are to be implemented. Next, the paper will describe how officers and politicians in Lund view such questions. Whilst outcomes have been different in Lund, their view is in many ways similar to the Eskilstuna politicians’ approach.

4.2. Long-term work for decreasing the share of travel by car in Lund through many small steps benefiting, walking, cycling and public transport

Lund is a city that for a long time has been working with decreasing the share of travel by car. Lund took important steps away from a car-centric planning as early as the end of the 1960s when the plans for a thoroughfare through Lund City Center were shelved (Lund Municipality, 2014, p. 3). Other important decisions that illustrate the city’s long-term work are when in the 1970s through traffic was stopped from directly crossing the city center and in 1999 when a transport plan was adopted that did not focus on the accessibility of car traffic but on walking, cycling and public transport, and which advocated mobility management measures (interview transport planning manager).

The interviewees highlight several working methods and factors as an explanation for their, relatively other Swedish municipalities, successful work to decrease the share of travel by car. The long history of planning for decreasing the share of travel by car may be an explanation for the fact that there is, according to the chairperson of the Building Committee, relatively broad and consistent political agreement on the transport policy goals. This political stability means that there are, according to both politicians and officers, no major changes in transport policy goals between election periods. The differences between the parties’ views on transport policy goals were described by a politician in the Technical Committee as ‘differences in the degree to which something should be implemented’ rather than fundamental disagreements about what should be implemented. The relatively high level of political agreement is most probably due to the city’s organizational culture that Hrelja (2015) described as deliberative and consensual, with working practices building consensus between politicians, and between them and officers, regarding objectives and transport and city development.

The officers’ understanding of the political conditions for the implementation of measures naturally presupposes that there is a dialogue with the politicians, which a politician in the Technical Committee describes as very important:

[…] we have a lot of dialogue between officers and politicians. In many municipalities there is a closed door and many officers are a little afraid to make contact with politicians, which means that all contact with politics must go through the managers, and that is very negative.

However, that is not the whole explanation. The City’s transport policy goals have also changed over time and they have developed from having in the early 2000s a focus on environmental aspects to also include economic and social aspects, i.e. all of the three dimensions of the concept of sustainability. According to a head of the Technical Services Department, the relatively broad political agreement on the transport policy goals can also be explained by the fact that with this shift in focus, it became politically easier for all parties to work for increased travel by public transport, walking and cycling when sustainable transport became part of an urban development policy that also contained goals for economic growth.

Despite this, measures to reduce travel by car can be controversial, even in Lund. Measures implemented in Lund that involve restrictions on car traffic, such as streets being converted into bus lanes and bumps being built on well-trafficked streets to reduce speeds and create safe bicycle crossings, are not understood as ‘sticks’ for car traffic but as ‘carrots’ for walking, cycling and public transport, and they are, according to the head of the Technical Services Department, communicated as such to the public. In addition, politicians rarely give the administration the task of implementing restrictions on car traffic. A transport planning manager emphasizes that:

We are rarely commissioned to reduce the competitiveness of car traffic, but [a reduction of the competitiveness of cars] usually happens in a context when we want to achieve something positive for walking, cycling and public transport. [Reduced competitiveness for car traffic] then becomes a natural consequence. It is rare that there is a political decision about implementing a deterioration [for car traffic]. It does not work politically. It is without a doubt on the carrot side that the majority of measures exist.

The interviewees emphasize that as an officer it is important to understand the political conditions for potentially controversial measures, here illustrated by the head of the Technical Services Department:

A good officer produces planning proposals that lead to decisions. You need a very great sensitivity about how far you can go and what the politicians who decide are mature to decide about. But at the same time, you must not lose professionalism. We must tell the truth, that there is a climate crisis. Now, we have to do more. But it is still the case that a decision must be made. It is not an easy task to be a politician.

A skilled officer thus is very sensitive to what is politically possible to implement and adapts the design of measures to what is politically feasible at that moment. The officers’ understanding for what is politically feasible gives an indication of the approach of implementing potentially controversial measures decreasing the share of travel by car. Instead of car restraining measures that can lead to rapid and radical reductions in car traffic, officers describe their work as a purposeful work for reduced travel by car, which through many small steps in the form of measures that benefit, walking, bicycle and public transport, all together over a longer period of time lead toward decreasing the share of travel by car, below described by a transport planning manager:

We do a lot but it takes time and we screw things up a bit all the time and try. Sudden or very large changes in direction are not easy to manage. We do some things well, we try to do them even better. Some things we do not do so well, and we try to turn them in a more sustainable direction. There are different areas you have to work with all the time to get there.

A concrete example of this strategy, which the officers themselves mention as a characteristic example of how they work to implement transport policy goals, is the planning of Ideon and Medicon Village. The area consists of two science parks and is Lund’s largest area for research and education. Since the beginning of the 2000s, the area has been developed with new workplaces, housing and services. The area is adjacent to European route E22, which is often affected by queues during rush hour traffic in northbound direction when there is a lot of traffic on the E22 that wish to reach Ideon and Medicon Village. At Ideon, the city is building a new motorway junction in collaboration with the Swedish Transport Administration to improve accessibility, mainly for car traffic. New road infrastructure that improves the accessibility of cars is the typical example of planning leading away from goals of reduced car traffic, and the road investment is also criticized by Green party politicians and NGOs.¹ What makes the officers interviewed see this as an example of a ‘good’ planning is that the construction of the new motorway junction is the last step in a much longer period of working with mobility management measures and measures improving walking, cycling and public transport in the area. Such measures have postponed the need for investment in new road infrastructure for almost two decades until it was no longer possible to continue developing the area without investing in new road infrastructure. Transport policy goals about decreasing the share of travel by car is not set against economic growth or urban development in this particular case, according to a transport planning manager:

If you were to say no [to the investment in road infrastructure], you would say no to a development of the area. You could imagine that you change society completely and in consequence of that say that we should not have any car traffic at all in this district, but we do not think it would be reasonable with the competition [between cities] if not all cities in the region do the same […]. So, you need to push towards sustainability goals as much as possible, but it must work within its context and reality. This is a weighing I need to do in my work, namely, how much can we push within what is politically feasible and acceptable? It may be difficult to understand. It is easier to say that a road development is wrong because it goes against sustainability goals.

According to this officer, what distinguishes Lund from other cities is that the new motorway junction in other cities would have been built from the very start. In Lund, the construction of the motorway junction is the end point of a long-term effort to postpone investments in new road infrastructure.

5. Discussion

Firstly in this discussion section, a brief summary of the findings is provided in relation to the first two of four research questions set out in Section 3, above. Thereafter, the findings are discussed in relation to the analytical framework based on path dependency, which also relate to the second pair of research questions. Finally, the findings from Lund and Eskilstuna will be discussed in comparison with previous research about other cities.

Lund and Eskilstuna have worked over the past ten years to make travel in the city less car dependent in a similar way but more has been implemented in Lund than in Eskilstuna. Similar objectives are set in both cities, to reduce carbon dioxide emissions from transport and to reduce levels of car travel. However, Lund has been able to deliver more in terms of integrated transport and land use planning to reduce the need to travel; parking management; and transformation of road space to space for other modes. These are the methods that have been used to reduce car travel, along with, in Lund, significantly increased public transport supply. In Eskilstuna, public transport supply has been increased but the transformation of road space and parking management have been difficult to deliver. An interesting similarity between the cities is the preference for pull measures over car restrictive push measures, and how the choice of measures is influenced by factors that prevent the implementation of radical car restrictive measures that could quickly reduce the proportion of car travelers.

The paper has sought to observe path dependency as a mechanism that limits or slows the adoption of more radical measures to restrain and limit car traffic, and to understand the mechanisms that produce it. It is clear that path dependency is present in both of the cities and city administrations studied. It is primarily caused by planning routines and norms in the public administrations, but the paper shows that these norms go beyond only the public officials themselves, and extend to the ways in which the officials and politicians interact. These are the institutions of path dependence identified by Low et al. (2005). These institutional path dependencies can foster sustainability transitions as in Lund or inhibit sustainability transitions as in Eskilstuna.

In Lund, the routines and norms support mutual understanding between politicians and officials; in Eskilstuna, the planning routines that define how the officials (try to) communicate with the politicians are often counterproductive. In addition, in Eskilstuna, structures, norms and routines that support car planning survive, whereas in Lund these have been supplanted by new norms – such as, for example, an acceptance of the need to try measures to manage traffic demand before building new infrastructure to cater for it – which make it more difficult to revert to strongly car-focused planning.

In terms of the physical factors that create path dependence, it is clear that Eskilstuna is still building infrastructure that supports and therefore ‘locks-in’ a car-dependent lifestyle, and the expectation that such a lifestyle will continue to be catered for. A good example of this is the large car park built at the city’s new arena. In Lund the opposite is the case: a focus on land use planning and in transport infrastructure on infrastructure for non-car modes gradually makes travel by car relatively less attractive.

The discourses that create path dependence are also observed in the two case studies. The ostensibly average mother, struggling to go swimming any other way than by car, cited by the Eskilstuna politician as a reason for building the car park is an example of a discourse of car-dependence. In Lund, a contrasting discourse is that of presenting sustainable transport as part of a coherent package together with economic and social sustainability, and gaining acceptance of that across political parties and the majority of citizens.

The paper has also sought to understand how transport planning path dependence in local authorities might be changed, so that the organization gradually shifts to a different, less car-dependent focus in transport planning. The case studied in this paper that has achieved greater change is Lund. Its change has been delivered in relatively small steps over a long period of time (since the end of the 1960s), although there were some major changes in infrastructure, such as the closure of the city center to through traffic, which were not possible to introduce gradually and so can be seen as a single major change. However, the account given by the interviewees from Lund is very much one of change in path dependency, and the creation of a new path dependency focused on planning for non-car modes, as the result of quite a long term and deliberative process of small steps.

The key in Lund was to create a common understanding of goals, measures and implementation among administrations and between politicians and officials through dialogue and understanding of the conditions that exist politically. The development of sustainable urban mobility planning in the city’s transport strategy, Lundamats – now in its third iteration – is one way in which these new norms and routines have gradually been built up over time. Lundamats provides a vision that the politicians are signed up to and for which they understand the rationale, and they are then likely to point out when a proposed measure does not conform to the strategy. This is a good example of creating, and then institutionalizing, structures that guide action, planning and decisions in a certain direction.

The result illustrates the need to see transport policy formulation and implementation as a phenomenon shaped by normative assumptions about what values are prioritized and what policy instruments are seen as appropriate or not in local planning contexts. Changing the discourse of transport planning has been important in Lund – the way in which car restraint measures are described and understood, as the other side of the coin of improvements to other modes of transport and to the city as a whole, is crucial. If the discourse, as in Eskilstuna, is one of car versus non-car modes, this conflict is less likely to deliver implementation of car restraint measures.

5.1. Lund and Eskilstuna’s experience in the wider context

The difficulties of reducing car traffic in Eskilstuna, in this paper explained as a result of path dependency, resembles the difficulties research report from other cities. Pflieger et al. (2009) look at the German city of Oldenburg and the French city of Clermont-Ferrand as cases of a continuation to the present day of traditional car-based predict and provide transport planning, due to an interaction between spatial planning and transport planning decisions that locked-in a car-based form of development, and due to very little innovation in leadership in these cities from 1945 to 1990. As they comment (p. 1430) regarding the German case, “Innovations which could have changed the trajectory of Greater Oldenburg systematically clashed with a very forceful reproduction dynamic” – meaning that transport and land use planning carried on much as it always had.

To succeed in breaking a path dependency, Low and Astle (2009) claim that: “Paths can be dislocated by courageous political leadership (for instance in Curitiba or Vancouver) or as a response to external pressures such as climate change or rapidly rising fuel prices.” However, they conclude that in their case study of Melbourne, no such changes have occurred and so transport planning and policy has continued on broadly the same roads-focused path over the past 50 years due to roads organizations being broadly stable in their form and strengthening over time in their access to resources; whilst public transport organizations have undergone a change in the opposite direction.

How, then, have cities that have managed to break path dependencies managed to do so? Pflieger et al. (2009) cite the cases of significant change in Karlsruhe and Basle where (p. 1432) “In both instances, the distinguishing features of their trajectories are the highly specific and contingent choices that reversed the trend in terms of transport-related decisions. These initial preferences could be consolidated and reproduced thanks to stable cognitive and institutional frameworks, and would ultimately generate what we can label as ‘innovation path dependencies’.” The ‘highly specific and contingent choices’ in the case of Karlsruhe were the decision in the 1960s not to close the city’s tramway system; and then in the 1980s the coincidence of three creative thinkers – senior members of staff in key organizations in the city who also were prepared to collaborate with each other – who invented the idea of tram-train as a solution to some of the perceived barriers to greater use of the regional public transport system.

Buehler et al. (2017) gives another example, Vienna. In Vienna path dependence has taken transport planning and policy in the direction of sustainable rather than car-focused transport and it now has one of the lowest levels of car dependence of any city in the developed world. At the end of WW2, the decision was taken to retain the tram system and (re-)build a dense and livable city, and this approach continued with decisions to start building the metro system in 1968, and the city’s first sustainable transport plan in 1980, and soon after that the implementation of parking management. The city has been able to maintain a very consistent transport policy because of the continual presence in local and national government of the Social Democrat party, but also because of a broad cross-party consensus on the direction transport policy should take. As Buehler et al. (2017 p. 268) state:

The implementation of sustainable transport policies in Vienna has been a long-term, multi-staged process requiring compromises, political deals, trial and error, and coalition-building among political parties and groups of stakeholders. This consensual approach to policy development has been very time consuming. […] Vienna’s risk-averse approach to adopting new transport policies has delayed their implementation but has increased the probability of success by learning from the experience of other cities. It has also tended to minimize political risk, and may help explain the Social Democrats’ continuous electoral success over the past seven decades, which has in turn facilitated the long-term development and implementation of transport policies.

There are clearly many parallels between the experiences of these other cities – both those that have adopted more sustainable policies and those that have not – and Eskilstuna and Lund. In the case of the former, the absence of a strong ‘break point’ or key event, or key person in power, is notable and so transport policy has continued in an uninterrupted direction with both institutional and discursive path dependence reinforcing each other, in a similar way as in Melbourne, Oldenburg and Clermont-Ferrand.

If previous research, on other cities as Melbourne, is used to explain path dependence in Eskilstuna, the absence of external pressure could be claimed to be one explaining factor. However, Lund is interesting insofar as the development of the institutional and discursive path dependence that take it in the direction of more sustainable and less car-focused transport does not appear to be associated with external pressures, or one particular break point or key event observed in other cities. Lund has not seen major changes like the tram-train in Karlsruhe, or the 1968 decision in Vienna to build a metro. Rather, there are a number of events over time that seem to have gradually changed the direction of the city’s transport policy: its closure of the city center to through traffic in the 1960s; a relatively environmentally aware and educated population; at certain times, key politicians, but not ones who endured in the way that those in Vienna did; and the adoption of a sustainable transport plan in 1999. In this respect its experience seems to have been somewhat different to that of other cities where change has occurred; but this does not mean that that change has in the end been less marked than that in cities such as Karlsruhe or Basle, for example. These differences contribute knowledge of those strategies that can be successfully pursued to implement measures to reduce car traffic, and in the following section we use these insights to discuss how the long term and deliberative process of small steps in Lund can be analytically understood from a strategy making perspective.

6. Conclusion and policy implications

Different policy approaches for how to manage and implement goals of less car use clearly exist between cities, in this paper illustrated by the comparative case study of transport and land use planning in two Swedish cities – one assessed by peers in the transport planning sector to be ‘leading’ (Lund), and one more ‘average’ (Eskilstuna). The differences in strategies and measures chosen arises because politicians and officers are influenced by a range of political, economic and administrative factors when trying to decrease the share of travel by car.

The paper shows that path dependency theory can provide an understanding of why certain types of transport (push) measures are not implemented in cities on as widespread a basis as policy objectives require. It appears that, based on the two cases examined here, change can be brought about by building new or modified institutions within the city administration, where routines and norms gradually change so that car restraint measures gradually become part of the normal way of doing transport planning. This then starts to lock-in certain patterns of travel and make further car restraint measures more feasible, and institutionalized as part of a standard menu of measures that the city uses, and not something out of the order.

In addition, changing discourses through carefully constructing a shared vision that places car restraint measures within a wider vision of how the future city will look and function is also critical. Over time, many small decisions in the form of measures that especially benefit, walking, bicycle and public transport together create a development direction that leads to realization of the goals of reduced car travel. In Lund, it is primarily these, pull measures, that interviewees claim are the most important measures to use when reducing the share of travel by car. The result has clear implications for the strategies cities should use to successfully implement measures to reduce car traffic. The result shows a clear dilemma. Extensive changes need to take place in travel habits if environmental goals are to be achieved, but the most effective strategies do not for all cities appear to be radical policies leading to fast implementation of goals about sustainable transport, for example by implementing very car restrictive measures, even in the face of resistance from the public and from within the city administration. Change through sudden large shifts in policy direction and hard car restrictive measures is not the way in which Lund has achieved its reduction in car use. This conclusion adds to what we know already about conditions for implementation, and casts doubt on the ability to make changes quickly in urban authorities where car restrictive measures challenge routinized practices embedded in power relations and discourses. The results support an approach that from a strategy making perspective can be understood as an institutionalizing process by which internal organizational and external public support for car restrictive and potentially controversial measures are built. This is a strategy that take a long time and even risk measures being watered down through negotiations, possible consensus-building procedures and compromises.

Disclosure statement

The authors declare no conflict of interest

Additional information

Funding

This research was funded by the Swedish Energy Agency (grant number 43202-1), which had no involvement in the study design, in the data collection, analysis, and interpretation, or in the writing of the paper.

Notes

1 See position of the Green Party: https://www.mp.se/lund/just-nu/miljopartiet-i-byggnadsnamnden-nej-till-trafikplats-ideon-ja-till-satsning-pa, and the Environmental Protection Agency Lund: Yttrande över Detaljplan för del av Östra Torn 27:2 m.fl. i Lund, Lunds kommun (Trafikplats Ideon), https://lund.naturskyddsforeningen.se/wp-content/uploads/sites/130/2018/11/Yttrande-Trafikplats-Ideon-detaljplan-granskning.pdf. Both documents accessed on 08 February 2021.

Appendix A.

Sources of data, Table 2

City populations: city authority websites, latest available year.

Mode share for cities: Lund Municipality (2018a), Eskilstuna Municipality (2016).

Car ownership/1000 people from: http://extra.lansstyrelsen.se/rus/SiteCollectionDocuments/Statistik%20och%20data/Korstrackor%20och%20bransleforbrukning/Tabell1.xlsx