Impediments for cycling
infrastructure funding
The case of Sweden
Emil Rensvala
November 2020
Supervisor: Peter Schmitt
Department of Human Geography Stockholm University
-1-
Rensvala, Emil (2020): Impediments for cycling infrastructure funding: The case of
Sweden
Urban and Regional Planning, advanced level, master thesis for master’s degree in Urban and Regional Planning, 30 ECTS credits
Supervisor: Peter Schmitt Language: English
Abstract
Increasing cycling and making it safer has been a policy at the national level in Swe-den for nearly two decades. However, the policy has yielded limited success. Insuf-ficient funding for safe and attractive bicycle infrastructure have been singled out as a key factor. The aim of this study is to examine the intersection where the policy at the national level in Sweden is to be translated into transport planning budgeting and the impediments for the allocation of funds. The method consists of analysing key texts and performing 28 interviews with government officials and national and regional planners to examine professionals’ views on budgeting practices. Actor-network theory (ANT) is applied for the analysis. Impediments are found in disap-pearing funds, responsibility-delegation to municipalities, unbeneficial cost-benefit methods and organisational hypocrisy. The national transport planning system, as it is enacted in terms of legislation, objectives and practices, is in some aspects rigged in disfavour of cycling investments.
Keywords: cycling, transport infrastructure, planning, policy, budgeting, national-level, Sweden.
The research took place in the Spring of 2017, and it has taken some time to finalise the report. The researcher bore at the time the last name Törnsten.
All quotations in the thesis originally written or spoken in Swedish are translated to English by the researcher.
Declaration of conflicting interests
The researcher was active in non-profit cycling advocacy at the beginning of the re-search period, and during the process started to work as a consultant for cycling advocacy organisations and later as an employee. This research has however not received any funding or been written on assignment.
-2-
Abstract ... 1
1. Introduction: Examining cycling policies at the national level in Sweden 4 1.1 ‘Increased and safe cycling’ – the Swedish national cycling policy ... 4
1.2 National-level cycling infrastructure funding – insufficient? ... 5
1.3 Aim and research questions ... 7
1.4 Demarcations ... 7
1.5 Empirical data collection ... 8
1.6 Theoretical approach ...10
1.7 Positionality and ethics ...11
2. Literature... 13
2.1 Budgeting and policy ...13
2.2 Budgeting for infrastructure and cycling ...14
2.3 Summarising the budgeting literature ...16
2.4 Cycling and budgeting in the Swedish context...16
3. Planning and budgeting in Sweden ... 19
3.1 Transport infrastructure in national-level policy-making ...19
3.2 Budgeting of transport infrastructure ...20
3.3 Available funds for bicycle infrastructure ...21
4. Disappearing funds ... 23
4.1 Pushed in time ...23
4.2 Long-term (dis)continuity ...24
4.3 Cut-offs...25
4.4 Lack of earmarked funds ...26
5. Allocating responsibility to municipalities ... 27
5.1 Displacing national responsibility ...27
5.2 Municipal funding of state roads ...28
5.3 Road legislation limits the building of state cycling paths ...29
5.4 Reshuffling interests and goals ...32
5.5 Lack of administrative champions ...34
6. Investments becoming unnecessarily expensive ... 36
6.1 Insufficiently examined cycling projects ...36
6.2 Planning and design demands ...36
6.3 (Un)common procurement and coordination ...38
7. Methods for prioritising investments ... 40
7.1 Deficient data ...40
7.2 Over-reliance on CBA ...41
7.3 Forecast planning ...43
8. Status, prestige and hypocrisy ... 45
8.1 Cycling as low status ...45
8.2 Cycling as non-prestigious ...46
8.3 Minimising the need for action ...47
9. Conclusion ... 50
9.1 What is the funding situation? ...50
-3-
9.3 How does planning processes impact funding? ...52
Implications for practice and further research ...53
List of interviews ... 55
-4-
1. Introduction: Examining cycling policies at the national
level in Sweden
1.1 ‘Increased and safe cycling’ – the Swedish national cycling policy
Swedish national-level cycling policy has been substantially examined by cycling policy advocate Spolander (2013b, 2013c, 2014b). In a famous photograph, he holds in his hands a pile of over 30 documents which he has analysed: strategies, action programmes, committee reports and political decisions in transport (see Spolander, 2014b, p. 7). Sweden has for years expressed willingness to achieve ‘increased and safe cycling’, the national cycling policy mantra.
All eight parties currently in the Riksdag (the Swedish parliament) are expressing the need for increased and safe cycling: for example, in the motions now to the
Riks-dag’s Committee on Transport (see Trafikutskottet, [forthcoming]). The
govern-ment (Social Democrats and Greens) recently produced a national cycling strategy (Regeringen, no date a). The political majority in the Committee on Transport ex-pressed the need as early as 1997 (Spolander 2013b). This evolved in the early 2000’s (Spolander, 2013b, 2013c, 2014b), and at least since 2008 (possibly earlier), all Riksdag parties, including opposition parties, expressed pro-cycling sentiments. In the right-wing government bill for transport in 2008 an entire section was de-voted to improve conditions for cycling, with infrastructure as the most important measure (Regeringen, 2008b), while cycling was simultaneously highlighted by all left and green opposition parties in motions (see Trafikutskottet, 2008). In 2010, the right-wing government requested a Swedish Government Official Report on cycling (Regeringen, 2010).
This is also articulated in many state reports of various kind: published by adminis-trative authorities or the ministries (see Spolander, 2013b, 2013c, 2014b). Several national cycling strategies(e.g. Vägverket, 2000; Trafikverket, 2011b, 2014), inves-tigations (e.g. Spolander and Dellensten, 2003; Vägverket, 2007; Trafikverket, 2011b; Johansson, 2012; WSP, 2012; Johansson, 2013; Trafikverket, 2016c; Daniels and Eklöf, 2017), and recommendations (e.g. Boverket et al., 2005; Faskunger, 2007, 2008b, 2008c, Statens Folkhälsoinstitut, 2009, 2010a, 2010b, 2011a, 2011b, 2013; Schantz, 2012; Trafikverket, 2010b) have been produced. The benefits of cycling are also well noted in these publications, and are also examined in other state-sanc-tioned reports (e.g. Faskunger, 2008a; Trafikverket, 2013; Johansson and Eklöf, 2015; Trafikverket, 2016c, 2016b; Schantz, 2016). Improved infrastructure is with-held as the key measure for increased and safe cycling.
In contrast, it is extremely rare to find statements that cycling should decrease or not be prioritised. What is less known is, of course, cycling’s relation to other modes of transport and the ambitions for such. A thorough academic study would be needed to identify what future transport situation that is really sought after. The long-term transport objectives were set out by the Social Democratic minority gov-ernment in 1998 (see Regeringen, 1998). It should be societal-economically effec-tive and sustainable while contributing to sustainable development and welfare for citizens and business. The Riksdag’s 2006 decided principles of national transport policies underlined that a range of alternative transports should be optional (see Regeringen, 2006, pp. 42–8). In 2009, under the right-wing majority government,
-5-
two equally weighted objectives were added; that of accessibility and equality, and that of consideration to safety, environment and health (see Regeringen, 2009a). It is fairly easy to position cycling within these objectives, but so but so can be ar-gued for other modes transport. The latest government bill on transport (prop. 2016/17:21), by the minority Social Democratic and Green Party coalition, men-tioned six societal challenges related to infrastructure: fossil fuel, construction of new housing, business, employment, digitalisation, and inclusion (Regeringen, 2016b). Cycling was particularly highlighted within the objective ‘continuing the promotion of effective and sustainable road transports’ (p. 51-2). ‘Increased and safe cycling’ has been repeated in politics and bureaucracy nationally since the turn of the millennium (see the above cited publications), but it has never been an official, national objective (although such has been suggested by e.g. Vägverket, 2000; Traf-ikverket, 2011b; Johansson, 2013; and Persson et al., 2015). For road safety, the Swedish ‘Vision Zero’, of erasing deaths and severe injuries in traffic, has guided the adjustment of the ‘road system’s design and function’ since 1997 (Regeringen, 1997). Of quantified objectives, there is one1 that directly relates to bicycle
infra-structure: 70 percent of large2 municipalities should be classified as having good
quality maintenance of prioritised cycling paths by 2020 (Trafikverket, 2015a). There is also a second indicator that is used: safe walking/cycling/scooter road crossings within urban areas, but there is no quantified objective attached to it (Ibid.). Trafikverket (2016a) suggested a 2020 objective to be 35 percent for this, and also to include state roads in this indicator. There are also a range of environ-mental3 and health4 objectives that directly and indirectly relates to cycling and
transports.
1.2 National-level cycling infrastructure funding – insufficient?
There is a new, annual national evaluation of cycling policy performance (Trafikver-ket, 2015d). Cycling policy has been examined in state-sanctioned investigations (e.g. Spolander and Dellensten, 2003; Vägverket, 2007; Trafikverket, 2011b; Johans-son, 2012; WSP, 2012), indicating room for improvement. In terms of policies for
increased cycling, the share is slightly below ten percent of all journeys, and has
re-mained so for the last decade (Spolander, 2013a; Trafikanalys, 2015; cf. Svensk Cykling, 2014; Trafikverket, 2015d), despite many cities reporting significant in-creases in cycling traffic (Trafikverket, 2010a). In terms of policies for safe cycling, cyclists are over-represented in Swedish traffic accident statistics. Cyclist’s risk of injury or demise has decreased, but not as fast as for cars, and it now represents the
1 There are ten indicators of road safety development. 2 More than 40 000 inhabitants.
3 The agenda 2030 objectives (Finansdepartementet, 2016), the national environmental quality
ob-jectives from 1999 (see Regeringen, 2001; cf. Naturvårdsverket, 2012), which since 2012 explicates cycle paths (Miljödepartementet, 2012, p. 102), and finally the objective that emission of greenhouse gases from transports should be close to zero by 2050 (see Regeringen, 2009b).
4 Better, and more equal, health of the population was a national level objective during the Social
Democratic reign until up until 2006 (see Regeringen, 2002). The promotion of cycling was explicitly mentioned (Ibid., p. 80). In the health politics of the succeeding right-wing majority government, the role of cycling infrastructure in promoting physical activity was more highlighted (see Regeringen, 2008a). This was repeated in the right-wing minority government’s announced health politics in 2012 (Regeringen, 2012). The current Social Democratic and Green Party minority coalition men-tions the connection between cycling and health in its latest budget bill (see Regeringen, 2016a, p. 45).
-6-
largest category among the severely injured (Trafikanalys, 2015). In 2015, only 25 percent of road crossings were considered safe, and 40 percent of municipalities were considered to have good maintenance of prioritized paths (Trafikverket, 2016a, pp. 49–53).
Determinants of bicycle use is complex, yet fairly researched matter. The overall quality and accessibility of the infrastructure (i.e. roads, parking, maintenance, sign-age, etc.) is often a key factor that influence the choice to cycle or not (e.g. Noland and Kunreuther, 1995; Pucher, 1997; Pucher and Dijkstra, 2003; Rietveld and Dan-iel, 2004; Moudon et al., 2005; Hunt and Abraham, 2007; Pucher and Buehler, 2007, 2008; Spolander, 2008; Akar and Clifton, 2009; Dill, 2009; Heinen, van Wee and Maat, 2010; Pucher, Dill and Handy, 2010; Buehler, 2012; Buehler and Pucher, 2012; Buehler and Dill, 2016; Harms, Bertolini and Brömmelstroet, 2016; Aldred et al., 2017; Brey et al., 2017; de Souza et al., 2017; Rojas López and Wong, 2017). Infra-structure is also key for avoiding injuries and deaths among cyclists (Reynolds et al., 2009; Niska and Eriksson, 2013; Trafikverkt, 2015e, Lindberg et al., 2016). Besides ‘hard’ infrastructure, ‘soft’ policies targeting the broader circumstances of transport and willingness to cycle contributes to ‘successful’ geographies of cycling (Pucher, 1997; Pucher and Dijkstra, 2003; Pucher and Buehler, 2007, 2008, 2009, Buehler and Pucher, 2009, 2011; Pucher, Dill and Handy, 2010; Gössling, 2013; Harms, Ber-tolini and Brömmelstroet, 2016; Buehler et al., 2017; cf. Trafikverket, 2012). Infra-structure in isolation may not be a guarantee for increased and safe cycling, but is perhaps the most important factor. And it needs funding. Insufficient funds have been singled out as a key factor in explaining the perceived lack of progress in cy-cling policy. Not only by cycy-cling advocates, but also official documents calls for in-creased state funding of both state and municipal bicycle infrastructure (Kågeson, 2003; Spolander and Dellensten, 2003; Vägverket, 2007; Trafikverket, 2011b; Jo-hansson, 2012; Spolander, 2013b, 2013c, 2014b; Persson et al., 2015; Trafikverket, 2016d; Lindberg et al., 2016, pp. 39–40).
This is also a view also amongst most of the fifty-one experts and practitioners who responded to the survey in the Swedish Government Official Report on cycling (see Johansson, 2012, pt. 2). Several of them were municipal officials, and such views have often been expressed (Lindberg et al., 2016, p. 39). Municipal officials claim that lacking resources, and continuity of such, is a major issue (WSP, 2012). Several ambitious cycling municipalities have jointly requested increased earmark funding (Styrelsen Svenska Cykelstäder, 2017)5. Also, in the preparations for the regional
plans for transport infrastructure, several regional actors expressed the funding need (Regeringen, 2016b, p. 69).
The costs for increasing cycling and making it safe was estimated up to 1 billion SEK6
“in the nearest years” (Johansson, 2012, p. 549). However, the inquiry makes no ref-erence to how this was calculated and the need is likely much higher. The Swedish Association of Local Authorities and Regions (SKR) estimated the mere costs of re-building 16 000 cycle passages in intersections across the country according to new legislation, to 3 billion SEK (see Johansson, 2012, p. 365). In 2007, Vägverket esti-mated the investment need to roughly 8 billion SEK for state roads (Vägverket, 2007). Spolander has estimated cycling investment needs to 15 billion SEK
5 Disclaimer: the researcher was not involved in Svenska Cykelstäder at the time. 6 10 SEK 1 €.
-7-
(Spolander, 2014b). The costs of fulfilling the investments of contemporary munic-ipal cycling strategies in Stockholm County is estimated to be 7,8 billion SEK (Envall, Johansson and Skiöld, 2015), and extrapolating this to the rest of the country leaves 40 billion SEK (Envall, 2016). An assessment made by Trafikverket (2016c) esti-mated costs for developing the transport system in accordance with a ‘climate sce-nario’ of decreasing fossil energy used in road transports with 80 percent until 2030, to 160-170 billion SEK. It suggested that at least 40 billion SEK should be earmarked funding in the form of grants for sustainable urban environments and public transport, cycling and walking.
1.3 Aim and research questions
The aim of this study is to examine the intersection where the policy for increased and safe cycling at the national level in Sweden is to be translated into transport planning budgeting and the impediments for the allocation of funds.
Three research questions have guided the research:
1. What is the current national-level funding situation for bicycle infrastruc-ture (in terms of what the existing funding streams are and how existing funds are utilised)?
2. What are the views amongst the involved stakeholders regarding funding responsibility (what should be funded by who)?
3. How does the tools and processes that are used in the transport planning impact on the distribution of funds?
‘Infrastructure’ in this thesis refers broadly to all the ‘hard’ measures that can be funded through public funding: building new roads and paths, improving existing ones, parking, signage, maintenance, and so on.
1.4 Demarcations
The thesis focuses on nation-state budgeting for cycling infrastructure (e.g. road construction, asphalt, parking, signage, maintenance, et cetera). This means solely policies related to infrastructure, and not on non-infrastructure policies concerned with the mobility of individuals. This exclude also funding of ‘soft’, mobility-related funding (particularly since those are funded through other streams than the infra-structure in Sweden).
Infrastructure in Sweden is also a municipal issue. As most cycling journeys are local (Trafikanalys, 2015, p. 17), it is often withheld by state officials as primarily a con-cern of municipalities. As to be seen, this discussion is in itself motivating this focus. County-specific geographies were not prominent in the empirical material.
Envall (2016) identifies three ways in which the need for cycling infrastructure in-vestments can be solved: through reprioritising funds amongst competing infra-structure investments or other expenditure areas; through increasing the amount of available funding; or through utilising available funds more effectively. As the the-sis is interested in how cycling infrastructure is constructed and contested, the ma-terial has been allowed to ‘speak for itself’ and being open-minded in relation to En-vall’s alternatives. However, to examine all three areas in detail would open the scope to beyond what is possible to scrutinise within the frame of this thesis. Envall’s
-8-
second area, to increase the overall transport infrastructure budget, would broaden the scope massively to include negotiations within the government of how funds are distributed between the many expenditure areas. The same to some extent applies to the third area, lowering costs, which is another topic that can be extended far into details of production technologies, et cetera.
1.5 Empirical data collection
Interviews with professionals were combined with text analysis. A sound starting point for a discussion on methodology, is on the ontology of the materials of the case, that is, what materials the researcher expects to find (cf. Whatmore, 2003). It was expected that those involved in the practice of balancing interests and prioritising that budgeting is, would be able to share their views of why funds were allocated as they were. It was needed to read the actual budgets, but it was also important to read other relevant texts to contextualise the interviews. ‘Texts’ contained digits, descriptions of digits, motivations for digits, statements, reports, laws and regula-tions, et cetera. Overall, the emphasis has been on words, artefacts, percepregula-tions, meaning, events, details, flux, explanations, and connections (cf. Bryman, 2012, pp. 380–414).
In selecting texts and respondents, the methods have emphasised flexibility and a limited structure (cf. Bryman, 2012, pp. 403–4). This means that the texts and re-spondents have been found and selected through ‘purposive sampling’ and ‘snow-ball sampling’ (Bryman, 2012, pp. 417–29), where texts and respondents made ref-erences to other texts and persons to interview. Texts and respondents were also found through informal conversations with a number of helpful researchers, experts and planners within the field.
The texts included for analysis was primarily reports, strategies and budget pro-duced by national-level administrations and the government. They were read in an unstructured manner to allow for flexibility to return to sections and in examining and extracting the most relevant bits. Texts were seen as ‘black boxes’: products of the budgeting processes that entailed segments of the story, or was referred to by other texts or respondents and thus brought ‘back to life’. The method was the often unspecified ‘qualitative content analysis’, which here was a broad searching for un-derlying themes (Bryman, 2012, p. 557).
Twenty-eight interviews were carried out with officials working with budgeting processes regarding state funding of transport infrastructure. Interviews were semi-structured and conducted in Swedish. Most interviews lasted some 30-60 minutes. The researcher listened and steered the direction of the interviews through more or less individualised interview guides, while the respondents did most of the talking. The form allowed both researcher and respondent to follow up on com-ments and ask for clarification. To a big extent, interviews revolved around ques-tions of “why?”, where the researcher generally explicated background and the hy-potheses while respondents reacted to it and discussed mechanisms. The setting of interviews varied greatly, as each was adjusted to the possibility to allocate time in the respondents’ often busy schedules. Thus, some interviews took place at lunch restaurants or at other locations where audio recording were inappropriate or oth-erwise difficult due to background noises. Transcription was in these cases instead based on notes, and took place directly after for the sake of remembrance. At two
-9-
occasions, two respondents were interviewed jointly to made the interview possi-ble. Many interviews were conducted on telephone, as the respondents worked in different parts of the country. These were audio recorded. There was little difference in responses gained in telephone respectively face-to-face interviews (cf. Bryman, 2012, p. 488), and both methods produced relevant material.
The interview guide below summarises the various topics which were talked about with respondents (with some variations):
Respondent background and cycling situation (e.g. a region’s decided fund-ing)
The general challenges for cycling funding.
The perceived willingness to invest in cycling.
Cycling not considered a mode of transport of its own – if so and how that is an issue.
Prestigious objects and how cycling investments are seen.
If and how cycling funding is disappearing.
Existence of strategic misrepresentation.
Attitude towards earmarked funding.
Processes of deciding objects, including cost- and benefit analyses, and con-sequences for cycling.
Co-funding practices and effects on cycling funding.
Distribution of responsibility.
Road and funding legislation
Cooperation with interest groups and other actors.
Seventeen were regional (county) infrastructure planners, working with producing the budget in county plans for regional transport infrastructure. They were asked about the structure of their county plans, their perception of the development of cycling funding and attitudes towards investments, how objects are prioritised and calculated, objectives and steering mechanisms (including legal frames), and coop-eration with road authorities (including co-funding) and NGO:s. A number of state-ments in the literature were specifically asked about, such as disappearing funds, ‘strategic misrepresentation’, earmarking funding, and diffusion of responsibility. This group was well represented, and the produced material seemed to fatigue to-wards the end.
Five interviews took place with current or former employees of Trafikverket, plus two regional infrastructure planners who were former employees at Trafikverket and could also express themselves as such. Interviews covered the same topics as with the county planners. The research faced difficulties in gaining access to re-spondents and some interviewees at Trafikverket and the Ministry of Enterprise and Innovation were reluctant to share their own reflections and were only willing to explain what could already be read in documents.
Three interviews were carried out with political officials of the two government par-ties, involved in national-level transport politics. Given the role in governing state agencies, deciding on laws, initiating national funding schemes, and specifying the demands on the national plan (where most funding exists) and county plans, as well as deciding on them, these interviews were far more relevant than interviews with
-10-
regional politicians. These interviews covered the governing parties’ intentions quite well as they were asked to explain their actions and to elaborate on the issues discussed in literature and interviews. One interview was carried at the Ministry of Enterprise and Innovation. Finally, two interviews were carried out with cycling ad-vocacy officials.
1.6 Theoretical approach
A budget regularly consists of two elements: digits, and written descriptions that act as explanations of the intended use of these digits. These intend to trigger certain actions within the state apparatus and, in turn, certain changes in space. In actor-network theory (ANT), a actor-network is a collective of associated humans and non-hu-mans. ANT may not be a theory (Mol, 2010), but a set of methodological principles about tracing and uncover relations and connections between actors (Bosco, 2006, p. 136). In this ontology, the world consists of networks of actors. Through studying and acknowledging the symmetrical potential of humans and non-humans to act to be and enacted, this ontology opposes the modernist division between society and nature that “renders invisible the political process by which the cosmos is collected in one liveable whole” (Latour, 1999, p. 304). Actors are active ‘mediators’ (Latour, 2005, pp. 37–42) that perform movements to allow this (Latour, 1999, p. 303). In this aspect, a budget can be seen as a ‘black box’, a ready-made product which, inde-pendent of its potentially controversial history and complexity is a given fact. ’Spokespersons’ speaks for black boxes (Latour, 1987, pp. 70–4). ANT travels ‘up-stream’ to uncover the construction of this black box (Latour, 1987). H – h
The concept of ‘translation of interests’ is of fundamental importance. For action to occur, the locking in of other actors is necessary (Callon, 1986). For a stakeholder to shape the budget according to its interests, it need to translate this need into actual digits. In this process, this interest is transformed in different way: becoming digits, and eventually a reshaping of space through infrastructure. Through this chain of translation, it interacts with a myriad of other actors and networks. As actors are drawn into a chain of translation, artefacts are created. A machine can only become successful if the right components are put in place and it is put into further practice. It is not that when a machine works that people will be convinced. Rather, a machine will work when all the relevant people are convinced (Latour, 1987, p. 10). This is how networks change, as they intend to become more sustainable in relation to other interests. In a translation, conventionally a word, it become equivalent but not exactly the same, and translation thus necessarily involves transformation (Law, 2009). This concept will allow the thesis to identify the development of the budget-ing process and the actors that were made important in construction the budget as such, as well as in the rationalities underlying its construction.
At the ‘centre of calculations’, where facts are brought together (Latour, 1987, pp. 215–257, 1999, p. 304), digits become ‘inscriptions’. For Latour, inscriptions refer to “transformations through which an entity becomes materialised into a sign, an archive, a document, a piece of paper, a trace” (Latour, 1999, p. 306). Thus, inscrip-tions are ‘matters of fact’: the outcome of a longer processes, which in turn can be further referenced, articulated and transformed while keeping some relations in-tact. Like maps of the Amazonia (cf. Latour, 1999, p. 29), budgets are necessary for orientation in the landscape of transport planning, without which everyone would be lost. Various instruments, or ‘inscription devices’ (Latour, 1987, p. 68), can be
-11-
used to make translations possible. Artefacts are loaded with meaning through what Latour (1999, pp. 49–51, 303) calls ’articulation’. The question is not whether it re-fer to a state of affairs or not, but how well it is articulated. To allow translations, events, places and people must be brought ‘home’ through being rendered mobile, being kept stable to disallow additional distortion, and being combinable “so that, whatever stuff they are made of, they can be cumulated, aggregated, or shuffled like a pack of cards” (Latour, 1987, p. 223). The quandary of a ‘fact-builder’ is to diffuse a fact in time and space (Ibid. p. 108): a fact, an argument or a thing must travel further and be used by others for its success.
A ‘proposition’ is for Latour (1999, p. 309) not a sentence that is true or false in the epistemological sense, but as what an actor offers other actors in the ontological sense. This is an important part of how translations occur. Latour (1987, pp. 108– 21) identified five strategies for an actor to translate interests: to enrol others, to be enrolled by others, to take a mutual detour, to reshuffle interests and goals (through five tactics: to displace goals, invent new goals, invent new groups, rendering the detour invisible, and, besides the primary mechanism of make collective action pos-sible, to win the secondary mechanism of being attributed a successful innovation), and, in sum, to become indispensable. A main programme of action might require ‘sub-programmes’, where successive movements are needed to reach the main ob-jective (Latour, 1996, p. 217). ‘Anti-programmes’ are programmes of action that are in conflict with the programme chosen as the point of departure for the analysis (Latour and Akrich, 1992, p. 261). Translations necessitates an object being ren-dered mobile, as “[a] thing can remain more durable and be transported farther and more quickly if it continues to undergo transformations at each stage of this long cascade” (Latour, 1999, p. 58). A ‘reference’ is an example of how a material object can be rendered mobile and exist in a thought or on a piece of paper through various transformations: “[i]t seems that reference is not simply the act of pointing or a way of keeping, on the outside, some material guarantee for the truth of a statement; ra-ther it is our way of keeping something constant through a series of transfor-mations” (Latour, 1999, p. 58). ‘Historicity’ (Latour, 1999, p. 306) is a term to high-light how time evolves through events and mediation, where actors are trans-formed.
‘Institutions’ provide all the mediations necessary for an actor to maintain a durable and sustainable substance (Latour, 1999, p. 307). After the enrolment of actors, in-terested groups must be controlled and kept stable to make action predictable through ‘machination of forces’ (Latour, 1987, p. 128). The successive and automo-tive movements of parts of a machine is an objecautomo-tive for a fact-builder, as depend-ency on other actors inherently means less control. Moreover, a chain is only as strong as its weakest link (Ibid., p. 121), which means that any component working improperly, whether in a product’s construction or diffusion, may mean its failure.
1.7 Positionality and ethics
Research is a two-way encounter affected by positionality. As Massey (2003, pp. 77– 80) argues, the researcher is constrained by discourse and data filtered existing frames of thought and language. Disinterestedness is one important aspect of ethics in research, and “means that the researcher must have no other motive for his or her research than a desire to contribute new knowledge” (Gustafsson, Hermerén and
-12-
Petterson, 2011, p. 70). This should be considered especially important when re-searching a topic closely bound to political priorities and (spatial) ideologies. The researcher is involved in interest groups for cycling and is thus neither distant from or disinterested in the conclusions from this research. This position has by large been positive for the execution of the research, since it has provided the researcher with insights and contacts. It has been important to disallow conclusions without scientific support, and not to withhold any material except for when it is ethically motivated. There has not been any interest of finding any particular results within the frame of the research questions.
Infrastructure planning consists of a small community and this motivated anonym-ity of respondents. Respondents could have a bias to exaggerate the negative impact of other actors and elevate their own role. To compensate for this, the study strived to interview ‘both sides’.
To ensure credibility, ‘respondent validation’ (Bryman, 2012, p. 390; Caretta, 2016) was used as quotations and/or my interpretations of the interviews were e-mailed to respondents let them correct these. Translation from Swedish to English were made with careful attention how things were said.
-13-
2. Literature
2.1 Budgeting and policy
For decades, it has been attempted to explain how public budgets are formed, and a number of theories of public budgeting in general exist. One group of normative the-ories concerns the overall purpose of the budget. These are, for example, the budget as a provider of collective goods, the budget as an instrument to counter economic fluctuations, the budget as provider of individual goods, the budget as promoter of effective programmes, and the budget as an instrument to maximise state income (Lane and Back, 1989, pp. 143–222).
Another group of theories have been developed to quantitively explain how succes-sive budgets develop in size and content. Two such theories of budgeting are domi-nant in political science (Breunig and Koski, 2009): incrementalism (hi occurrence of small-scale shifts in politics) (cf. Bäck and Larsson, 2008, p. 294), and punctuated equilibrium (occasions of dramatic change). In the incremental theory, last year’s budget is the largest determining factor of the size and content of upcoming year’s budget (Mortensen, 2005). Budgets can be characterized by both stability and punc-tuations (Mortensen, 2005). These studies are often quantitative, and focus on size and content rather than the decision-making process (e.g. Jones and Baumgartner, 2005; Mortensen, 2005)
Within this quantitative literature, there are also studies of that have attempted to explain why budgets change. In studying the reasons for budget change in American state politics, Breunig and Koski (2009) found that the governor’s role was central (see also Barrilleaux and Berkman, 2003). A study by Jacoby and Schneider (2001) found that budgets varied across US states in the same way as interest group partic-ipation and public opinion. However, in a Danish study, newspapers had little impact on local budgeting (Mortensen and Serritzlew, 2006). Epp et al (2014) examined partisanship and budgets and found that it had little impact on budgeting. Rather, inconsistent allocations occurred at almost exactly same rate as consistent. The ex-istence of ‘political budget cycles’ is also one area that has been subject of research to verify if and how expenditure is related to the electoral cycle (e.g. Bojar, 2017). A third group of literatures have instead taken interest in the processes that shape budgets. A multitude of theoretic approaches have been deployed. In incremental theory, budgeting decisions evolve in a limited social system (parliament, govern-ment, authorities, etc.) of clearly identified roles of those who demand funds and those who decide on expenditures. Neither have full information of alternatives and consequences, and cannot act rationally (Lane and Back, 1989, pp. 172–86). Thus, this pragmatic theory argues that the implementation of policy is more likely when more people are ‘on board’ (Allmendinger, 2009, p. 142). In one such study, Cova-leski and Dirsmith (1983) discovered that budgeting is not a process where top lev-els controls middle and lower tiers of the organisation. Rather, budgeting is a pro-cess with an upward flow of advocacy.
This approach is closely linked to new institutionalist theories. These are united by their conviction that institutions (customs and rules that provide the structures, context, stability, and constraints for interactions) matter (Dodds, 2013, pp. 231–8).
-14-
A distinct theory is the ‘garbage-can’ theory. In argues that decisions are made based insufficient information, inconsistency, inability to connect means to ends, and changing rules and participants. Problems, solutions, participants and choice oppor-tunities are mixed in an irrational process (Lane and Back, 1989, pp. 194–6; Bäck and Larsson, 2008, pp. 294–5; Dodds, 2013, p. 242), where decisions are shaped by bounded rationality (Eisenhardt and Zbaracki, 1992).
Another study of budgeting within this group of literatures is by Nils Brunsson. For Brunsson (2007, ch. 5), budgeting can have many effects on organisations; it can for example be at the heart of policy processes, have a symbolical function, act as an instrument of control, be a source of negotiations, and as a final product turn out quite different from what was first expected. Brunsson discovered different roles in the budgeting process: those who championed various operations, those who were guardians of the cash box, and those who were hoarders who argued for stockpiling revenues. The size of the budget was shaped by a struggle between stakeholders regarding access to information and construction of responsibility. Moreover, allo-cated money does not necessitate action: “the coupling between budget decisions and action is a very loose one” (2007, p. 70). The theory of ‘hypocrisy’ (Brunsson, 1986, 1993, 2007) states that political talk, decisions and actions can take different directions. For Brunsson (2007, p. 112), “to talk is one thing; to decide is a second; to act is yet a third […] People may talk or decide about a certain action but act in the opposite way. The result is hypocrisy.” Inconsistencies occur. Conflicting exter-nal demands can be satisfied through inconsistencies, and lack of action can quiet opposition through talk. To talk or decide things can be as valuable for organisations in modern society as actions, since publicity matters. This is the primary driver, but a factor can also be that some issues can only be talked about, decided or acted upon, and not vice versa. The model assumes that consistency in all three levels will create dissatisfaction. Thus, the more something is talked about, the less likely is action, since talk about one area must be compensated with decisions and actions in other areas, and vice versa.
An alternative, closely linked perception of budgeting departs from rational choice and public choice theories. Individuals and authorities act according to self-deter-mined preferences (Dodds, 2013, pp. 190–3). Authorities will apply for more fund-ing than needed, and they also have monopoly on their services. This results in so-cietal-economic ineffectiveness (Lane and Back, 1989, pp. 200–7). The significance of this approach is, like the above, that budgeting consists of a multitude of interests. However, in empirical testing, it is not necessarily so that too much funding is pro-vided. For example, Hagen (1997) studied budgeting from a principal-agent per-spective, and found that the principal, in this case counties in Norway, had a firm position in relation to the agent (hospitals).
2.2 Budgeting for infrastructure and cycling
In transport studies, research on the ‘economic’ dimension has mostly concentrated on economic instruments to manage travel behaviour, such as taxation (Schwanen, Banister and Anable, 2011, p. 995). While there are small threads in various aca-demic and non-acaaca-demic reports that help to give insight into factors that may affect funding, resource allocation is occasionally noted or described in transport policy research (e.g. Marsden and Rye, 2010, pp. 673, 674), but rarely investigated. One such example is Mahendra et al (2013), who notes that there is a mismatch between
-15-
objectives and resource allocation in India, where most resources are invested in infrastructure for motorized vehicles. In analysing the transport budget for the In-dian city Pune, Menon (2011) found a similar mismatch. Makarewicz et al (2010) contrasted national transport policies and funding shares in the UK, Canada, Sweden and the USA. Like for political science approaches to budgeting as described above, there exist studies of how infrastructure budgeting change in overall size and con-tent (e.g. Legacy, 2016). In Global North geographies, scholars have identified a shift from the public provision of infrastructure into more market and individually fi-nanced and managed infrastructure projects (Graham, 2000; Koppenjan and En-serink, 2009; O’Brien and Pike, 2015; Strickland, 2016).
For cycling, there is a broad range of academic literature that notes a lack of funding, suggest increased funding to improve conditions for cycling, or highlight the role of funding in best-practice (e.g. Pucher and Dijkstra, 2003; Rietveld and Daniel, 2004; Moudon et al., 2005; Pucher and Buehler, 2007, 2008, Buehler and Pucher, 2009, 2011; Cradock et al., 2009; Pucher, Dill and Handy, 2010; Knight-Lenihan, 2013; Gössling and Choi, 2015; Buehler et al., 2017). Determinants of bicycle infrastruc-ture funding are largely unaccounted for. In Ireland, clientelism has been noted in distribution of cycling funds (Manton, 2016). In New Zealand, Knight-Lenihan (2013) could identify how legislation affected funding for sustainable transporta-tion. There, discursive storylines also shape budgeting practices (Imran and Pearce, 2015). In Canada, local planners felt that funding was determined by different ob-jectives at different governmental levels (Hatzopoulou and Miller, 2008). Cradock et
al (2009) quantitively examined funding shares across the USA, and found that
counties characterized by persistent poverty or low educational status were less likely to fund cycling.
In the UK, funding has been doubly devolved; away from the state, and away from transport, according to Aldred (2012). Funding mostly appeared on the local level, as well as being sourced from non-transport budgets such as for schemes in health (Aldred, 2012). However, the ‘English Cycling Demonstration Towns’ was a joint na-tional-local funded investment programme (Department for Transport, 2010). The UK Department for Transport recently announced a Cycling and Walking Invest-ment Strategy with earmarked funding (DepartInvest-ment for Transport, 2017). Such na-tional funding of local bicycle infrastructure is common in many countries, such as the Netherlands, Denmark, Germany (Rietveld and Daniel, 2004; Pucher and Bueh-ler, 2007, 2008, Buehler and Pucher, 2009, 2011; Pucher, Dill and Handy, 2010), the USA (Cradock et al., 2009), New Zealand (Knight-Lenihan, 2013), Norway and Swe-den; it is generally the size of allocated funds that differ. Moreover, in the UK, the central government adjusted funding allocations up or down based on quality of lo-cal plans and achievement against the targets proposed in these plans (Marsden, Kelly and Nellthorp, 2009). In the USA, the federal level was the one which increased funding for cycling, which did not occur in a similar rate at state level (Pucher and Buehler, 2006). In Canada, funding was almost exclusively local (Ibid.). But in the USA, other financing mechanisms than government funding are also common (Riggs and McDade, 2016).
-16-
2.3 Summarising the budgeting literature
In sum, little is known of budgeting and prioritising funds within the context of moting sustainable transportation. Budgeting in general is a complex set of pro-cesses where political and administrative concerns come together to shape the out-put. Infrastructure budgets are also shaped by how a topic such as cycling is enacted in relation to problems, solutions, responsibility, and so on. Yet, the end product is not a guarantee for anything to happen in the next step.
Studies have primarily focused on the overall size of budgets and on the political struggle of increased expenditures versus keeping expenditures low. Although gen-eral theories exist that can explain how certain policy areas are translated into budg-ets, there are limited empirical accounts of this. One exception to this is a study on how the defence budget decreased in Argentina (Pion-Berlin, 1998). It found that the shift to democracy shifted also the institutional structure. As budgeting become more institutionalised and less individualised, politicians saw less reasons to inter-fere.
There are also limited accounts of the role of non-human actors in the budgeting processes. Quantitative approaches have formed a substantial share of research in the area in, and qualitative approaches have focused on strategies employed by var-ious human actors (e.g. politicians and authorities). How legislation, documents, ob-jectives, et cetera, shapes budgets, is largely unknown. One exception here is a qual-itative study by Albert-Roulhac (1998), which analysed how national and suprana-tional budgeting process became increasingly hybridized, which happened due to, amongst other things, regulations.
Third, budget processes are virtually unstudied in geography, planning and transport studies.
2.4 Cycling and budgeting in the Swedish context
Some directions of areas to explore can be identified in various academic and ‘grey’ literature related to cycling and budgeting in the Swedish case as well as in some more general theories of planning and politics that does not fit into the ‘budgeting’ strand examined above.
A first point of departure concerns the construction of bicycle funds. It has been claimed in several non-academic reports (Kågeson, 2009; Spolander, 2013b, 2014a, 2014b) that allocated funds tends to disappear. A highlighted reason is that funds are concentrated towards the latter half of the budget period and thus can be more flexible. The disappearing can perhaps be caused by, first, earmarking of funds for older, ongoing projects, (see Trafikverket, 2016c, p. 48). Second, continuity in fund-ing has been highlighted as an issue (WSP, 2012). Third, cyclfund-ing funds might be sub-ject to reallocation to cover cost-overruns in other prosub-jects. That cycling paths are the ones first cut off from funding is noted in a European context (Buehler and Pucher, 2009, p. 51) as well as a Swedish (e.g. Spolander, 2013b). Flyvbjerg (2009) have suggested deliberate miscalculations; ‘strategic misrepresentation’, as a rea-son for cost-overruns. Fourth, in light of requests for earmarked funds for cycling (e.g. Vägverket, 2007; Kågeson, 2009; WSP, 2011; Spolander, 2014b), it is of interest to enquire how this is anticipated.
-17-
A second strand is about allocation of responsibility. In politics, responsibility for unwanted actions are best avoided through not executing them, making someone else seem responsible, hiding them, or reframing them as positive (Brunsson, 2007 ch. 5). First, how responsibility is constructed is of interest, not only for revealing how actors approach cycling, but specifically for identifying if lack of action in in-creasing funding is connected to unwanted responsibility. Second, this also includes how legal frames and planning practices distributes responsibility. The 2000 Na-tional Cycling Strategy underlined the need for a clear-cut division of responsibility when it comes to cycling paths (Vägverket, 2000). Since about a decade (Trafikver-ket and SKL, 2011), an increasing number of regions have a principle that munici-palities should co-fund cycling paths along state roads, and has been criticised by Spolander (2013b, 2014b). Third and fourth, defendants of investments could or could not perhaps be found within the administration (Brunsson, 2007, ch. 5) or in objectives, assignments, and strategic development plans at national and regional levels (Spolander, 2006; Kågeson, 2009).
A third area to explore is the cost of bicycle infrastructure and if it is lessening in-vestments incentives. That cycling interventions are insufficiently examined and are faced with cost-overruns can be one reason (Kågeson, 2009). A second issue is if design demands are causing interventions to be costly. It has been suggested that design manuals need improvements to better accommodate cycling (WSP, 2011; Jo-hansson, 2012). Third, it has been suggested that costs arise due to lack of common procurement (Envall, 2016) and increased coordination efforts have been requested (e.g. Vägverket, 2000). Fourth, it has been put into light that cycling paths need to be located where there is a [motor] road (e.g. Olsérius, 2016). Potentially, this can in-crease costs, and so can other aspects of planning processes.
A fourth point of departure concerns the role of methods used to prioritise projects over others. The Swedish Government Official Report on cycling (SOU 2012:70 [Jo-hansson, 2012, p. 155]) express the need to develop calculation principles for cy-cling. This is related to road data and data of need for investments, which are two areas where there has been lacking knowledge (Spolander, 2006; Vägverket, 2007). The issue was noted already before the 2004-2015 planning period (Kågeson, 2003), and before the 2010-2021 period (Kågeson, 2009). Related to this is lacking knowledge of shares and flows of cyclists (Vägverket, 2000, 2007; Regeringen, no date a; Trafikverket, 2013). Moreover cost-benefit analyses (CBA) has potential to be developed to better incorporate benefits of cycling (van Wee and Börjesson, 2015). The need to develop calculation principles for cycling has been expressed by
Vägverket (2000, 2007), Trafikverket (Schantz, 2016, pp. 56–8), SIKA (2008),
Jo-hansson (2012, p. 155), Spolander (2013c), the government (Regeringen, 2016b, p. 27, no date a, p. 17), and in academic literature (e.g. van Wee and Börjesson, 2015). Johansson and Rosander (2017) have found that such analyses is given highly val-ued in planning. A final issue here is the role of forecasts in planning (Johansson and Eklöf, 2015, p. 71).
A final topic is how the willingness to invest is enacted. First, the traffic engineer was a key actor in the production of a ‘car society’ in Sweden in the mid 1900’s, ‘techni-fying’ (de-politicising) a political question (Lundin, 2005). In planning theory, the role of planners as advocates and political actors is increasingly acknowledged in research (Allmendinger, 2009, ch. 7). Amongst transport infrastructure planners,
-18-
studies have identified a path dependency in that professional groups are character-ized by knowledge and skills aiming at optimizing the flow of car traffic (e.g. Aretun and Robertson, 2013). Second, it has been acknowledged in earlier studies (e.g. Flyvbjerg, 2014) that politics tends to concentrate investments to prestigious pro-jects. If this is the case, it can be a reason why little funding is left for cycling. Third, it has been concluded that the legal framework for safe and increased cycling gen-erally exists, but that there is a lack of willingness to improve it (Johansson, 2012). This is where politics enter. In politics, stakeholders struggle to shape the agenda of the public debate, as which issues exist at the forefront of the public debate (and which not), and ‘how’ they are discussed (and not), tend to pressurise politicians in making certain decisions (cf. Bachrach and Baratz, 1962). Therefore, it is important to enquire how cycling infrastructure is constructed in the political debate. A theory related to this that of ‘hypocrisy’ examined above.
-19-
3. Planning and budgeting in Sweden
3.1 Transport infrastructure in national-level policy-making
The provision of road and parking infrastructure in Sweden is largely provided by public funds and free to use (Hasselgren, 2013b; Dillén and Haapaniemi, 2017). The central state plays a key role in shaping infrastructure through funding, legislation, policy directions and executing those.
Roads are according § 1 Road Act (1971:948) (see Näringsdepartementet RS T, 1971) open for public traffic, with the state or a municipality as road authority. Pub-lic roads in Sweden total some 8 000 km of state trunk roads, 90 000 km of other state roads, and 41 000 km of municipal roads (Trafikverket, 2017a, Trafikverket Official 1). Some 430 000 km are private roads (Trafikverket, 2017a).
National level policy is produced by the national government and the Riksdag (the Swedish national parliament). The government decides on decrees, such as instruc-tions to state agencies (e.g. Trafikverket) on their purposes, while the Riksdag de-cides on ordinary laws and constitutions, and controls the government (Bäck and Larsson, 2008, pp. 66, 84). Decisions in the Riksdag are in practice most often taken in the committees (Ibid., p. 82). Infrastructure is primarily discussed in the Commit-tee on Traffic (Trafikutskottet).
The Swedish government is the primary institution for suggesting law changes and the state budget, and in executing the Riksdag’s decisions (Bäck and Larsson, 2008, p. 178). It currently consists of a minority Social Democrat and Green Party coalition. The Minister for Infrastructure Anna Johansson, Social Democrat, is responsible for these questions but formally, the government make decisions collectively (Ibid., pp. 178-81). Minsters have, besides their political staff, an administrative staff at the ministries (Ibid., pp. 181-8). Infrastructure is currently dealt with by the Ministry of Enterprise and Innovation (Näringsdepartementet), although cycling has been in-volved in policies at Ministry of the Environment and Energy (Miljö- och
energide-partementet) and the Ministry of Health and Social Affairs (Socialdeenergide-partementet).
These shapes much of the content in the government’s politics, including bills, the budget, Riksdag interpellation answers, and administrate matters.
National administrative authorities carry out the government’s politics through bal-ancing three principles: according to the political majority, to legislation, and to ef-fectiveness (Bäck and Larsson, 2008, p. 195). These are controlled through six means: 1) legislation, such as ordinance containing terms of reference for the au-thority, 2) funding, such as letters of appropriation, 3) recruitment, through control-ling leading positions in the authorities, 4) form control, through controlcontrol-ling the form of the authorities, as well as information and education, 5) board members, which are elected by the government, and 6) informal contacts (Ibid., pp. 198-206). The Swedish Transport Administration (Trafikverket) is such an administrate au-thority7. According to § 1 SFS 2010:185 (see Näringsdepartementet RS T, 2010),
Trafikverket is responsible for long-term planning infrastructure planning for all
-20-
modes of transport and for constructing and operating state roads and rail. Moreo-ver, it should on the basis of spatial planning create conditions for a societal-eco-nomically effective, internationally competitive and long-term sustainable transport system, and strive towards the transport-political objectives. It budgets the 12-year National Plan for the Transport System (hereafter ‘the national plan’) (Näringsdepartementet RS T, 2009a). There are also other state administrative au-thorities that are related to infrastructure matters.
County administrative boards carry out and coordinate the government’s politics at the regional level in the 21 counties of Sweden (Bäck and Larsson, 2008, pp. 196, 207–8). County councils are another regional authority that follow the same borders as the county administrative boards, but are self-governed by regionally elected pol-iticians, and thus cater regional interests. The distribution of tasks between county administrative boards and county councils differ. Some county councils are entitled ‘regions’, and have taken over responsibilities of regional development and infra-structure from the county administrative boards (Ibid., pp. 233-6). In some counties, there are other municipal cooperation organisations at the regional level which have the responsibility for regional development (Ibid.). The island Gotland is a special case, with a combined municipal-regional authority. In each county, either of these three regional organisations establish a binding County Plan for Regional Transport Infrastructure (hereafter ‘county plans’) (Näringsdepartementet RS T, 1997, 2009b).
3.2 Budgeting of transport infrastructure
At the beginning of the annual state budget process (for all expenditure areas), the government receives various allocation proposals, analyses and accounts. A budget bill is proposed during the spring, and in the autumn a definite budget proposal. Eventually, the Riksdag decides on the budget in December after discussions in the different committees (Petersson, 2007, p. 165, 2017, pp. 106–8; Bäck and Larsson, 2008, pp. 88–90).
The long-term economic road transport infrastructure planning is manifested in 12-year plans, renewed every four 12-years. The current period is 2014-2015, and the up-coming period 2018-2029 is under preparation. The national plan budgets invest-ments in construction and maintenance on trunk roads (stamvägnät) and grants for maintenance of private roads (Näringsdepartementet RS T, 2009a). Measures esti-mated to cost more than 50 million SEK (100 million in the coming period) are named and put on a list, (hereafter: ‘named-and-listed’) (Regeringen, 2017c). The government ultimately decides on the plan. Similarly, regional authorities (see above) in the 21 counties produces county plans (Näringsdepartementet RS T, 1997). Of relevance here, these plans budget funds for non-trunk state roads and co-funding of non-state infrastructure8. Measures over 25 million SEK are
named-and-listed (Ibid.). Trafikverket execute the plans.
The 4-year planning process is performed in two steps: orientation planning
(inrik-tningsplanering) and measure planning (åtgärdsplanering), which are repeated
every fourth years in preparing the renewed 12-year National and County plans
8 E.g. measures for improved environment and traffic safety on municipal roads (e.g. cycling paths)
(Näringsdepartementet RS T, 2009b). However, state grants. such as grants for sustainable urban environments [stadsmiljöavtal], are not funded through the county pns.
-21-
(Wennberg et al., 2011, pp. 19–30). The orientation planning centres around inquir-ies of the strategic development of the transport system, based on a government mandate (Ibid.). When the overall budget frames are set by the Riksdag, the govern-ment provides a new mandate to establish the measure planning, which is to budget the plans in detail (Ibid.).
Besides the national and county plans, there is a third way in which the state fund transport infrastructure: grants. Since 2017, municipalities and county councils can apply for national cycling co-funding through ‘grants for sustainable urban environ-ments’ (stadsmiljöavtal), if they in return do measures for increased shares of sus-tainable transport or more housing (Miljö- och energidepartementet, 2015). An-other grant is the ‘Klimatklivet’ [grants for climate investments], managed by
Natur-vårdsverket. Funds are distributed to a broad range of actors for measures that
mit-igate emissions (Naturvårdsverket, 2016). Thirdly, the county plans can also co-fund municipal infrastructure.
3.3 Available funds for bicycle infrastructure
In the current planning period (2014-2025), the overall frame is 522 billion SEK (Trafikverket, no date b). Of these, 281 billion SEK is for development of the transport system, 155 billion SEK is for maintenance and improvement of state roads and grants for private roads, and the remaining 86 billion SEK is for mainte-nance and improvement of state rails (Ibid.). Most of this sum is for the national plan: only 35,4 billion SEK is for all the county plans. In the upcoming period (2018-2029), the total infrastructure budget will increase to 622,5 billion SEK (Regeringen, 2016b). The county plans will get 36,6 billion SEK (Regeringen, 2017c, app. 1 p. 21). Exactly how much money that is spent on cycling in Sweden at any level is unknown (Vägverket, 2007, Trafikverket official 1). For named-and-listed interventions that contain cycling, but where the key object is another (often a road for motor traffic or parking in public transport hubs), cycling funding cannot be identified separately (Trafikverket Official 1). This is the case for both the national and regional plans. The issue was perhaps even greater some years ago (Kågeson, 2003). What can be identified, is the funding in the named-and-listed projects that are primarily for cy-cling, as well as pools for cycling. However, these projects and these pools are in most cases also for walking, and in some cases for public transport and other inter-ventions. Cycling funding can therefore only be estimated.
In 2000, cycling was estimated to make up one percent of the total national spending on transport infrastructure (Vägverket, 2000, p. 64). The government currently es-timates total bicycle funding in the current planning period (2014-2025) to be 4,8 billion SEK or around 0,9 percent of the total frame (Regeringen, no date a, p. 8). In the national plan, there are two named-and-listed cycle paths, totalling 0,16 billion SEK (Trafikverket, no date a), one of which was stopped recently for being too ex-pensive (see more in ch. 6). The government estimates total cycling funding in the national plan to be 1,4 billion SEK, and 3,4 billion SEK in the county plans, or around 10 percent of the county plans’ total sum (Regeringen, no date a, p. 8; see also Spolander, 2014a, 2014b; Svensk Cykling, 2014). In average over the three most re-cent years, Trafikverket have spent 136 million SEK on newly built walking and cy-cling paths (Trafikverket, no date d, table 38). Simultaneously, 16 million SEK have been spent on levelled intersections for walking, cycling and mopeds (Ibid.).
-22-
Of the mentioned state grants, the sum of funds for cycling depends on granted ap-plications. Total available funding for grants for sustainable urban environments in 2017 is 750 million SEK (Regeringen, 2017b). Some 700 million SEK9 per year is
available for grants for climate investments (Naturvårdsverket, 2016). In 2016, for the first time10, the government earmarked 25 million SEK for cycling in
Trafikver-ket’s letter of appropriation (Regeringen, 2016c). The money was extended in 2017
to 75 million SEK (Regeringen, 2017b). Funds should be used for attractive and safe cycle paths along both trunk and other state roads (Ibid.). In the ‘Sweden negotia-tions’ (Sverigeförhandlingen), a multi-level negotiation programme to increase housing through major infrastructure investments (Regeringen, 2014b, 2014a), the state will earmark 233 million SEK for cycling investments in the Skåne and Stock-holm regions (Sverigeförhandlingen, 2015, 2017b, 2017a).
9 See also Regeringen (2017a). 10 At least back to 2010, likely longer.
-23-
4. Disappearing funds
The tendency of cycle funding to disappear in Swedish national funding is claimed in several reports, as seen in section 2.4. This chapter verifies that it can happen and uncovers its mechanisms.
4.1 Pushed in time
When something is further away in the plan it’s easier to move it […] They’re not decided upon in the same way. It’s much more difficult to move an object closer in time. (Regional Infrastructure Planner 12)
Cycling funding can be concentrated towards the end of the budgeted planning pe-riod (Spolander, 2014a). Projects decided years ago are delaying investments in cy-cling:
We have a plan now where there are strong [cycling] priorities further in the period […] We don’t reinvent the new plan. The foundation is… at least the first six years are pretty steered. The first four or five [years] are in principle mort-gaged. (Regional Infrastructure Planner 9)
The many years it takes for projects to be planned and constructed locks funding for a long time. Most of the 281 billion SEK for development of the transport system in the current planning period (2014-2025) are already earmarked projects starting in the first years, which means that only some twelve percent are available for new projects within this time period (Trafikverket, 2016c, p. 48). As noted by a
Trafikver-ket official (no. 5), “we live now with decisions taken long ago”.
The institution (Latour, 1999, p. 307) that provide all the mediations necessary for this network to maintain a durable and sustainable substance is dependent on a number of translations. A first translation is that named-and-listed projects seldom are primarily for cycling, as expressed by a regional planner (no. 3): ”it’s difficult to find cycling paths of that size”. Many regional planners speak of cycling as a measure belonging in the non-named-and-listed section, as does the main rapport from
Traf-ikverket’s orientation planning (Trafikverket, 2015b). In the analyses provided,
there is no mentioning of the strategic need to develop the cycling path network, as there is for other modes of transport. It thus kept off from developing into a costlier undertaking.
A second translation is to fill up budgets with named-and-listed objects. Such objects “have an upper hand in the plans” (Regional Infrastructure Planner 9). The govern-ment’s directives for the measure planning (Regeringen, 2017c) offers a proposition (Latour, 1999, p. 309) in this case: orders are to finish already started projects, and that projects starting in the plans’ early years are investigated properly. This mobi-lisation makes budgets resistant to policy punctuations. To halt initiated projects is difficult: “we can’t stop an object in our county plan, then I’ll have to inform our pol-iticians that it has become more expensive and they’ll have to accept [it]” (Regional Infrastructure Planner 9).
-24-
A third translation is to view total frames as too small, causing little funding to be “left over” for cycling after other modes of transport have been prioritised. Several regional planners (no. 3; 4; 8; 9; 10; 12; 17) have complaints about the small frame of the county plans. One regional planner speaks of lack of leeway:
We, who doesn’t have a vast economic space, we can fit one major object, one can say. That is – one major road project. If one is unfortunate it will take a considerable share of the plan. […] It’s problematic that we don’t have that much money to use. One doesn’t have leeway, in some way. I note that quite big parts of the plan are locked on certain measures that I know will be exe-cuted. […] Then one will have to use the remaining millions as best as possible, and distribute them. (Regional Infrastructure Planner 8)
However, there are also examples of anti-programmes (Latour and Akrich, 1992, p. 261). One region has opposed disappearing funds through a policy that funds in pools cannot be moved to other pools (a sort of earmarking), and managed in that way to secure allocating funds (Regional Infrastructure Planner 10).
4.2 Long-term (dis)continuity
Many feel [the grants] should be more long-term. (Regional Infrastructure Planner 7)
Continuity in funding could lock up funds for cycling over several years. But conti-nuity has been highlighted as an issue, which hampers cooperation between munic-ipalities (WSP, 2012) and common procurement. The different time frames of mu-nicipal and state plans create an uncertainty in co-funding processes. While national and county plans range for twelve years, municipal budgets often have shorter time spans. This means that municipalities are facing difficulties in living up to their long-term undertakings:
Some municipalities have applied for and received funds, but […] have not done what they promised to do […] At a guess, one has been compelled to reprioritise in the budget for different reasons, and there hasn’t been money left to do this. (Regional Infrastructure Planner 17)
In this quote, the regional planner underlines how lack of available funding in com-bination with their flexibility translates into discontinuity. Another planner empha-sises the importance of coordination:
We are definitely not alone to such a coordination problem. And one of the reasons is […] that municipal plans often are year. […] Even if one has a 1-year budget, one has a plan for the next 1-year. But [the issue] is to […] share these plans. That the right people are part in this, it can be about that as well – that the right people are part in the measure process. So of course, when it comes to state co-funding, we have a problem, and I’m sure you’ve heard of it – that they [municipalities] apply for funding but doesn’t use it. It’s difficult to control unless one is demanding, but also to have multi-years planning. But it also has to be that the processes are better synced. (Regional Infrastructure Planner 16)
