Shaping smart cities: problem framing, vertical selection and governance in UK smart cities

ABSTRACT As urban governments adopt smart city strategies for delivering services, the need to understand how – and in whose interests – these strategies are formed is imperative. The selection of smart city verticals (or areas of focus for smart city programs) within processes of urban governance has implications for which aspects of the urban agenda become prioritized. Through a study of seven UK smart cities, the paper investigates the framing of city problems, selection of smart verticals, and decision-making logics. The findings highlight that the selection of smart city verticals within the case study cities is rooted in four key considerations: challenges in service delivery, pragmatism, entrepreneurialism, and broader national and global events and policy agendas. These considerations transcend different spatial scales and governance arrangements, raising questions around democratic accountability and transparency. The study concludes that caution is warranted when framing smart cities as a solution to city problems.


Introduction
The aftermath of the 2008 financial crisis and austerity policies pursued by many national governments have left many local governments in a weakened fiscal position amidst growing demand for municipal services (Cardullo et al., 2019).Hence, in a search for cost-effective and efficient means of service deliverycombined with a drive to appear innovative in order to attract investmentmany city leaders have turned to smart cities, often framing them as solution to pressing urban problems (Cook & Valdez, 2021;Karvonen et al., 2019).While smart cities have grown in popularity over the past two decades, the verticals (or areas of focus) in which smart city initiatives sit and how they are selected remains an under researched issue.
While we return to these verticals and explain some of the common areas of focus for smart city initiatives in the literature review, we maintain that interrogating how smart city verticals emerge in the first instance is important for at least two crucial reasons.First, Hollands (2015) reminds us that "the smart city is made of a diverse range of things and can inadvertently bring together different aspects of urban life that do not necessarily belong together, hiding some things and bringing others to the ideological fore" (p.64).This implies that smart cities can downplay or overlook some urban problems while privileging others.This has important implications for shaping circulations of power within a city, and for shaping the governance network itself, as Söderström et al. (2014) summarizing Callon (1940) observe that a "crucial step in the creation of sociotechnical networks is the problematization of a situation in order to become indispensable actors in a network of solutions" (p.309).Which issues become prioritizedand in whose interestsand which become overlooked has clear implications for who benefits from the smart city, and for how urban inequalities are tackled or entrenched.
Second, in most urban areas, the provision and management of some urban infrastructures and city services fall within the purview of multiple tiers of local government (Karvonen et al., 2019), as is the case in the United Kingdom where waste management falls within the remit of county councils, and city or district councils are responsible for rubbish collection (GOV.UK, n.d.).This means that depending on the area of focus chosen, the selection of verticals could have profound implications for urban governance, in terms of the locus of decision-making and modes of demanding transparency and accountability.
By following this line of inquiry through interviews with individuals involved in smart city governance across seven UK cities (Oxford, Milton Keynes, Nottingham, Leeds, Newcastle upon Tyne, Bristol and Birmingham), we seek to shift the ongoing discussions about smart cities to the decision-making process involved in the selection of their verticals.This is important, as unpacking the roots of vertical selection in local governance networks can help to illuminate why smart city initiatives might or might not be able to address the city problems which are often used as a justification for their relevance.The paper also highlights that the selection of smart city verticals as an aspect of smart city governance is not distinct from the politics and complexities that characterize well-established aspects of urban governance at the local level.
To this end, the paper is organized as follows.In section "Literature review", we review the literature on smart cities, considering their driving forces, common smart city verticals and some of the initiatives which fall within each category.We ground the selection of verticals in a discussion of urban governance.In section "Methodology", we outline the research strategy, setting out the justification for the case studies chosen for the research and the methodology and approach to analysis.We present our research findings in section "Findings" and discuss their implications and conclude in section "Discussion and Conclusion".

Smart cities: definitions and discourses
Since the 1990s, smart cities have been understood to encompass "technology-based innovation in the planning, development and operational management of cities" (Tan & Taeihagh, 2020, p. 146).However, there is a lack of consensus around precisely how a smart city should be defined.Vanolo (2014, p. 884) observes that the smart city "is an evocative slogan lacking a well-defined conceptual core," thus allowing different "communities of interests" (including funders, academics, consultants, policymakers) (see Pascavela et al., 2015) and a "well-organised epistemic community" (see Cardullo et al., 2019) "to use the term in ways that support their agenda" (Vanolo, 2014, p. 884).
Meanwhile, Kitchin (2016) suggests that smart cities can be understood from three main perspectives.The first is about how the digital instrumentation of cities can influence the configuration and management of urban infrastructures.The second perspective views smart cities as those which deploy initiatives aimed at strengthening policy-making, and urban governance through the use of information technologies.The third perspective sees the smart city as the application of digital technologies and ICT to advocate a citizen-oriented approach to urban development and management which places emphasis on accountability, innovation, social justice, and transparency (Kitchin, 2016).
Critical commentators draw attention to how the city is conceived or framed at both ideological and normative levels (Green, 2019), the types of publics conceived (Joss et al., 2019), the form of governance enacted (Cardullo & Kitchin, 2019;Leleux & Webster, 2018), the nature of citizen participation (Shelton et al., 2015) and the ethical issues underpinning their governance (Ehwi et al., 2022).For example, Cardullo et al. (2019) observe that the smart city frames the city as a system rather than a place and it takes a technological solutionist approach to problem solving.
Commenting on the role of discourses in 5,553 smart cities globally, Joss et al. ( 2019) suggest that they "shape concepts and programs and constitute key avenues through which ideas and practices are borrowed, transmitted, and reproduced within different geographical, cultural and institutional settings" (pp.4-5) as they purposefully recast cities from places with histories and lived experiences to ephemeral notions of projects, hubs, laboratories, platforms and urban experiments.Cook and Valdez (2022) also argued that "as smart technologies are embedded in the urban fabric, technical standards become de facto law and technical expertise acquires unintended political powers" (p.3).Similarly, Söderström et al. (2014) observed that smart cities create "new relations between technology and society" as the latter becomes increasingly dependent on the former for solutions.As part of this process, "discourse around smart cities become a tool to make certain actors and technologies 'obligatory passage points' or key actors in the development and implementation of specific forms of urban management solutions" (p.309).

Smart city verticals
Verticalsor areas of focuscan be thought of as the lens through which city problems are seen and the specific aspect of cities that smart city solutions engage with.Hämäläinen (2020) adds that emphasizing verticals such as transportation or energy influences the type of technologies and standards that are chosen to support the needs and requirements of a particular market or industry.Although "verticals" are a familiar concept in the business and management literature (see Harrrigan, 1984), the smart city literature has appropriated it as a shorthand for conceiving the city, allowing city managers to engage with the markets in terms of buying into new partnerships, expertise, funding arrangements, infrastructure and standards (c.f.Angelidou, 2015).
Smart city verticals emerge from an instrumental perspective of the city and its problems.Among instrumentalists, comprising but not limited to academic institutions (e.g.universities and research centers), industry (e.g.technology and finance) and governments (at national, regional and city levels) (see Mattern, 2019) a central argument is that a city comprises its physical and social infrastructure through which it produces the public sphere (Joss et al., 2019), and therefore the embedding of technological devices within existing urban infrastructure, and the subsequent impacts on different aspects of social life, warrant attention in understandings of smart cities (Batty, 2012;Joss et al., 2019).From this orientation, the city is problematized in a way whereby urban questions are turned into "engineering problems" which can be measured, analysed and solved using quantitative methods.This instrumental view is also supported by a narrative for a radical transformation of how cities are managed following the "big data revolution" and the disruptions caused by breakthroughs in new technologies (Joss et al., 2019).
The material manifestation of this perspective is often seen in the application of digital technologies to existing city infrastructures to solve pressing urban problems (Kitchin, 2016;Söderström et al., 2014).Against this backdrop, several studies have attempted to categorize the areas of focus for different smart city initiatives.For instance, Kitchin (2016) suggests the technologies deployed in smart cities fall under eight domains (or verticals), comprising: (1) Government (e.g.online transactions and city operating systems); (2) Security and emergency services (e.g.centralized control rooms); (3) Transport (e.g.intelligent transport systems), ( 4) Energy (e.g.smart grids); ( 5) Waste (e.g.compactor bins), ( 6) Environment (e.g.sensor networks for capturing data on pollution); ( 7) Buildings (e.g.Building management systems), and ( 8) Homes (e.g.smart meters).
As such, it is evident that the categories of smart city verticals identified vary and sometimes overlap.However, there is a need to consider how these domains come to be and the governance and decision-making processes underpinning their selection.We therefore draw upon theoretical insights from the urban governance literature to interrogate decisions made regarding the choice of smart city verticals.

Smart cities and urban governance
In seeking to examine how areas of focus for smart initiatives emerge, it is pertinent to consider as a key line of inquiry how urban governance operates in the process of vertical selection.Processes of framing, in which particular options are constructed as more desirable than others, are essential to understanding how verticals are chosen for smart city development.As Callon et al. (2009) highlight, while technical issuessuch as the disposal of nuclear waste, or in our case, the selection of smart city initiatives to address urban problemsare often framed by apparently scientific actors as having an obvious or necessary technical solution, recognition of a multitude of actors highlights that there exists considerable debate on socio-technical issues, and decision-making on such issues is inherently political.Indeed, Farías and Blok (2016) argue that urban power is increasingly to be located within technical infrastructure systems, including data platforms, and that "government has become a techno-political art entailing the design, configuration, maintenance and repair of the infrastructures we live by" (p.540).The processes by which verticals for smart city development are chosen are therefore revealing of the ways in which urban power operates, and of the crucial moments of decision-making which shape outcomes for urban space.
Several trends in urban governance (in particular, neoliberal urban governance) are of relevance in the debates around smart city management.Smart cities themselves have arisen in the context of a protracted period of austerity and continually diminishing local authority budgets in the United Kingdom, overseen by successive coalition and Conservative governments, which have led many local authorities to seek efficiencyenhancing measures (Hastings et al., 2015).This is a widespread trend, spanning several countries and numerous urban municipalities.In a study of Italian smart cities, Pollio (2016) highlights that while smart city technologies have often hidden urban policy decisions under a veil of technical solutions to addressing urban problems and managing austerity-induced budget cuts, they are far from apolitical.Indeed, smart initiatives represent a political tool aiming to deliver on values of efficiency, and to replace funding-intensive urban projects (which are often impossible within constrained local authority budgets) with a series of short-term experiments in addressing urban challenges, which risks overlooking the long-term structural issues which underlie such challenges (Cowley & Caprotti, 2019).
Importantly, the emergence of smart urbanism has been identified as a key development in neoliberal urban governance: Grossi and Pianezzi (2017) argue that the geographical imaginary of the "utopia" represented by much of the rhetoric surrounding the notion of smart cities strongly reflects a neoliberal agenda.They highlight that the increased reliance of local authorities on private sources of funding, and on service-delivery mechanisms which involve public-private partnershipsall of which are part and parcel of many of the smart cities hailed as being for the public goodare characteristic of the neoliberal tendency to exalt inter-urban competition and market-led approaches to dealing with urban problems.Yet while smart cities may be relatively new vehicles of urban neoliberalism, many of the techniques and strategies of urban governance entailed have been well established for several decades.
Entrepreneurialism as a governance approach has long been recognized as part of urban neoliberalism.In the 1980s, Harvey (1989) identified a shift in focus of local urban governance (including not only formal organizations of local government but also a host of other actors) from providing services, to attracting private investment to stimulate local economic growth.Indeed, Harvey (1989) noted that this is a phenomenon which is rooted in transformations in capitalist development which have seen widespread deindustrialization in advanced economies, alongside heightened structural inequalities, and that the subsequent shift towards an entrepreneurial mode of urban governance represents an effort to engage in inter-urban competition for resources in what is framed as a zero-sum game where those cities that fail to effectively compete for capital lose out to more entrepreneurial cities. Jessop and Sum (2000) define entrepreneurial cities as those which develop and deliver strategies intended to ensure that they remain economically competitive (or become increasingly so).Critically, entrepreneurial cities explicitly project themselves as being entrepreneurial, and include "entrepreneurialism" as part of the narration of their identity (Jessop & Sum, 2000).It is easy, therefore, to see how smart cities can in many instances be understood to be entrepreneurial, given that they are often portrayed as centers of innovation, and may be viewed as a key means of attracting investment as a result (see Cardullo et al., 2018).
Nonetheless, the entrepreneurial smart city represents a departure of sorts from earlier iterations of entrepreneurial urbanism: McGuirk et al. (2021) argue that smart cities have seen a shift away from the enabling role traditionally played by the entrepreneurial local state, which occurs as organizations of local governance react to the perceived need to prioritize economic growth.They highlight that alongside the characteristic enabling of private investment (often entailing local state absorption of risk) and speculation, entrepreneurial governance in smart cities adds experimental practices (which come with inherent risk) to the mix, and works to actively constitute smart cities through these practices.And while this entrepreneurial governance in smart cities is often aimed at growing a network of knowledge-economy start-ups (see Rossi, 2016), and fostering competitiveness on an interurban scale, McGuirk et al. (2021) argue that these are not the only arenas in which "active, constitutive and experimental practices" (p.1744) play out, with other aspects of urban governance, focused on non-economic facets of city life and the day-to-day running of services, also undergoing changes as a result of these practices.
Studying smart cities therefore offers an opportunity to examine nuances within trends in urban governance (see Rossi, 2016).Peck et al. (2013) argue that while neoliberalism is a hegemonic ideology, it is not fixed, and continually evolves in localized settings, where specific local historical contexts shape the conditions of its manifestation.Citiesand indeed, smart citiesare not simply imposed upon by the processes associated with neoliberalism, but are key sites for its continual reiteration (Peck et al., 2013).As such, the remainder of this paper seeks to interrogate the specific ways in which neoliberal urban governance occurs across a number of UK cities implementing smart initiatives, and to show how these governance processes shape the landscape of smart initiatives which emerges within and across these settings.
As this paper considers how city problems are framed and verticals (or areas of focus) chosen for smart initiatives within cities, it is worth considering here who makes these choices and how.Decision-making within smart cities has an ethical dimension, whether or not this is explicitly recognized by those making decisions (Ehwi et al., 2022).Decisions are often made with reference to a system of norms and values, both personal and organizational, which shape the ethical reasoning carried out by decisionmakers (see Hunt & Vitell, 2006).As such, the norms and values which filter through an organization are critical in shaping decisions, and given that neoliberal ideology remains hegemonic in urban governance (even if there are local differences in its manifestation), it is possible to trace the ways in which neoliberal values underpin choices in smart city governance.The remainder of this paper considers in depth how specific verticals become the focus of smart city initiatives in cities across the United Kingdom, and considers what the processes of vertical selection can reveal about urban governance.

Methodology
To interrogate how smart city verticals become selected through processes of decisionmaking in urban governance, the United Kingdom was chosen as the national context for this research.This is partly because the United Kingdom government (via the Department of Business, Innovation and Skills) has funded several smart city initiatives to position the United Kingdom as a leading exporter of smart city consultancy and services (Cardullo et al., 2019).Additionally, UK cities such as London, Manchester and Newcastle have featured in both global and European Smart City rankings (Manville et al., 2012).Thus, the United Kingdom provides an appropriate context to draw relevant insights about smart city development generally.However, there is a practical challenge, in that there is no official list of UK smart cities and their respective initiatives.So, we began our inquiry by reviewing existing reports that have documented smart city initiatives across the United Kingdom, and found the report on smart city demonstrators by Farías & Blok, 2016, to be among the most comprehensive in terms of breadth.Next, we created a database of the demonstrator projects captured in the Future Cities Catapult report to create a chart of different smart city project demonstrators and the verticals they sit in (see Figure 1).
Following from Figure 1, we shortlisted cities that have implemented multiple verticals (at least four verticals), to allow us to engage with questions regarding the selection of the different verticals and their governance.As the catapult report was slightly dated, we visited the websites linked to the smart city demonstrator projects, including websites of their respective city councils to ascertain the status of the reported demonstrator projects.We decided to focus on cities which had smart projects that were still ongoing or had been scaled up.
Local authority officials from seven cities comprising Oxford, Milton Keynes, Nottingham, Leeds, Newcastle, Bristol, and Birmingham agreed to participate in the study.Semistructured interviews were held with a total of 16 participants comprising 10 council officers managing ongoing smart city initiatives in the respective councils, four elected politicians who sit on committees with remits on economic development and urban regeneration, a technology consultant and an academic working closely with councils on smart city projects.
The interviews took place between June and September 2021 and were organized via Microsoft Teams video calls.The interviews explored questions regarding the problems interview participants believed their cities were facing, the decision-making process regarding the choice of city problems to focus smart initiatives on, the governance and management of smart city verticals and the challenges that come with them, and the ethical considerations regarding the choice of smart city verticals.The interviews were recorded with permission from the participants and later transcribed.
In terms of analysis, the approach to thematic analysis suggested by Braun and Clarke (2012) was adopted.Both authors independently examined interview transcripts to become "intimately familiar with the data sets" (p.61).Next, initial codes were generated from the extracts at both descriptive and semantic levels.The authors independently developed themes around identification and framing of city problems, and the factors influencing the selection of smart city verticals and a meeting was later held to compare and refine common themes and also reconcile divergent themes based on criteria such as the depth of the theme, its coherence and evidence base.This peer-review process of data analysis has been found to counteract "sole-author biases" whilst improving the robustness and confidence in the empirical findings reported (see Adzimah-Alade et al., 2020).The authors then named the themes by paying attention to how they come together to elucidate how smart city areas of focus or verticals emerge.The final stage entailed combining descriptions, arguments and inferences to answer the research questions whilst embedding the analysis in the existing literature on smart city governance.

Findings
The case study cities The cities we studied differed in important respects, including their population, UK region, local governance arrangements, and gross weekly pay (See Table 1).These differences have a bearing on the types of problems cities face and the smart city initiatives deployed.However, there are also common features among the cities.For example, cities with larger populations (above half a million people) namely, Birmingham, Leeds and Newcastle are found within metropolitan areas.Similarly, cities in Southern England, namely, Oxford, Milton Keynes and Bristol have higher gross weekly pay (above £650) than their counterparts in the Midlands and Northern England.These similarities suggest that some problems might be common across different cities.

City problems and their framing
The list of problems cities face is potentially inexhaustive.However, what city stakeholders choose to focus on in their communication is important because it can provide impetus for selecting particular smart city initiatives as well as justification for the decisions made.For the case study cities, interview participants mentioned some city problems much more frequently than others.For example, transportation-related problems (comprising traffic congestion, high carbon emission, ageing transport infrastructure, single person car use, etc.) and poverty-related problems (including incidence of deprivation, social inequalities, child and adult poverty, and street begging, etc.) were among the most frequently cited issues.These were followed by housing-related problems (including unaffordability, homelessness, fuel poverty, and energy inefficient buildings).Other issues highlighted included healthcare and connectivity-related challenges.
Besides the identification, how stakeholders framed and talked about the problems facing their cities were instructive.While some problems were framed as "actually existing" (Brenner & Theodore, 2002) or "existent problems" (Dillon, 1982), others were framed as "potential" problems (Dillon, 1982).These two approaches to framing city problems are significant because with the former, interviewees reasoned that such problems affect or are experienced by the general public on daily basis.Hence, such problems are not difficult to prioritize in terms of gaining public support and investment.
Transport is the top problem … it always comes up.-Council officer (illustration of an "actually existing" problem) However, in the identification of potential problems, foresight is required in order to appreciate the need to invest so as to stay ahead of the curve.Interview participants suggested that unlike existent problems, potential problems were difficult to justify both to the public and potentially to political leadership: The original plan was [for] a city for 200,000 people, we're now at 280,000, there are plans to grow it to between 450,000 and 500,000 if you include the Greater Milton Keynes area … So, what the Smart Cities stuff is trying to do is effectively get ahead of the curve on some of these challenges, not necessarily the challenges today, but trying to plain and smooth the challenges before it becomes gridlock.And that has drawbacks because, when you're doing things, when people aren't [yet] sitting in a traffic jam for 25 min, the case for change is much more difficult.-Council officer (illustration of a "potential problem") Arguably, because potential city problems and their case for change are not immediately obvious, including to the public and political leadership, the process of selecting these problems as foci for smart initiatives is not likely to be bottom-up or co-creative (Hollands, 2015).This itself creates potential challenges for urban governance, as relying on technocratic insights and assumption to ascertain city problems risks detaching city problems selected for smart city investment from the reality of city residents' everyday lives (Goodman et al., 2020).It is also clear that decisions around which city problems to prioritize are not devoid of departmental politics and subjectivities: So I suppose what I'm going to say I think are the city's priorities, and I won't choose highways, but if you probably spoke to somebody in [the department for] highways, they would say "ours is much more important than the other stuff because without highways or without transport planning, you wouldn't be able to do [x, y or z]".-Council officer.
From the foregoing, it could be argued that that the issues that gain attention as city problems and how are they are framed are neither accidental nor passive (see Cook & Valdez, 2021;Joss et al., 2019).Rather, they are carefully framed so that they can effectively rival other issues.Indeed, given that funding sources are limited, the push for some problems to be given attention could relegate other issues to a subsidiary position.There is a clear ethical dimension to such decision-making, wherein imbibed norms and values constitute a strong guide in the policies pursued at local level (Ehwi et al., 2022).

Ongoing smart verticals
It is worth briefly discussing here the smart city initiatives happening across the case study cities, and some of the verticals in which they fall.Several smart initiatives are focused on transportation, including e-scooter trials, starship delivery robots, electric bus fleets, and EV charging ports, etc. Next generation connectivity, comprising projects such as the installation of wireless networks in public buildings, and initiatives aimed at addressing digital exclusion, is the second most common vertical across the case study cities.Other verticals included those focused on energy, assistive living, and governance.
Our analysis suggests that, although the prominence of the smart mobility vertical matches with the popularity of transportation-related problems, this is not the case for all verticals.This does not only point to a misalignment between city problems identified and the smart city initiatives deployed, but goes to show how technical solutions cannot be used for some city problems as the quote below illustrates: … inequalities are not caused by a lack of technology.They're caused by a whole heap of things further upstream.[For instance], racial disparities … will probably not be solved by technology, you could only mitigate it, but it's not solving an issue -Council officer There is a recognition that some problems are complex and cannot be solved by technology, with mitigation appearing to be the only option.However, within the context of digital innovation, some interview participants suggested that complex problems can be tackled through simplification.However, such simplification risks missing the bigger picture, and developing a focus on a narrow aspect of a complex problem.
If you're going to try and solve a problem … you've got to simplify it … with the complexity in mind.And I think also if you're going to develop an innovation, you have to think of the niche.You're not going to solve every problem -Interview participant.
From an ethical decision-making perspective, questions arise over which aspects of the complex city problem are sacrificed or downplayed to arrive at niche areas, which will in turn support technical solutions.Indeed, the remainder of this paper focuses on how particular verticals come to be selected, rather than on these verticals themselves.

Factors underpinning the selection of smart city verticals
The study found that four broad considerations, including challenges with the delivery of city services, pragmatism, entrepreneurialism, and existing global and national imperatives underpin the selection of smart city verticals.Note that there are some overlaps in the factors identified, and we do not claim that these are the only drivers of vertical selection.Instead, the four factors which emerged from the research findings provide a useful tool for analysing the different ways in which smart initiatives emerge, and for demonstrating the variable ways in which smart city governance operates to produce different smart city agendas in different cities.

Vertical selection based on existing city services
Interview participants suggested that smart city verticals often emerge as a result of challenges identified within the delivery of existing city services.For example, in Newcastle, some residents reportedly expressed dissatisfaction with waste collection services.The council invested in smart bins which helped improve waste collection by ensuring that bins that were full were prioritized during the collection process, eliminating instances of missed bin collections and reducing the need for refuse vehicles to routinely service areas where waste bins are nearly empty.Similarly in Leeds, challenges existed around delivering health care services, as patients' health records were not harmonized.The council thus made health care a key part of its smart city vertical by leveraging its existing close working collaboration with the National Health Service (NHS) Trust in the city and data hub to ensure that there is responsible data sharing among recognized healthcare providers.Reflecting on the strides the city has been making in smart health, a councillor from Leeds City Council observed that their smart health vertical enables improved datasharing: [It] allows you to get access to your health records now so that these are more transportable and transferable.So that if you go to a different health setting, it's a lot easier for your data to be available to someone who's making a decision, whereas in the past … for hospital appointments, [people] had big files in about three or four different departments.And they never ever consolidated these files together.So nobody ever knew whether or not there were any conflicts in [a person's] health records -Councillor, Leeds Four reasons explain why local authorities establish a smart city vertical around existing services.The first is that it helps local authorities to directly discharge their statutory responsibilities (e.g.waste management and adult social care) for which taxpayers are charged (Cséfalvay, 2011).Second, it allows councils to use their existing city assets for their smart agendas, without the need for investment in new assets.Third, it ensures that councils can leverage their existing experiences and expertise in specific aspects of service delivery to help improve the service.Finally, it helps the councils to retain significant influence over decision-making and project implementation, especially when the rollout of the smart initiative involves partnerships with third parties.

Vertical selection based on pragmatism
City governments often also take a pragmatic approach to vertical selection, choosing initiatives which draw on their existing strengths.This was the case in the selection of verticals in Oxford.Being globally renowned for its higher education offering, as well as several technology start-ups, the city leveraged the strength of its intellectual capital in the beginning of its Smart Oxford Programme.Officials explained that the program started as conversations between university academics, city officials and third sector organizations and resulted in a stock-taking exercise of the range of digital initiatives already happening across the city which could be brought together under the Smart City umbrella.The stock-taking revealed that there were several initiatives happening around mobility, and hence it was pragmatic to feature mobility as a key anchor of the smart city program.
Officials from Nottingham pointed out that city leadership also played an important role in not just the selection of smart city initiatives, but overall smart city strategy: I think it first of all comes down to who's driving technological revolution within councils and services … I think if you're within the homelessness department, for example, and whilst you're absolutely committed to service improvement, you might not naturally go to that innovation space, certainly around technology as much as another employee.So I think it's historically the types of conversations that happen within those departments, the type of staff, the type of mindset, etc.One of the challenges we have in [the Smart City team] is unlocking that creative thought process across service areas -Council officer, Nottingham This quote suggests that the type of leadership within the local authority, the organizational culture across different departments and the dominant mindset and conversations that take place within council departments hold immense sway in determining which smart vertical to include in the smart city agenda or exclude.An implication of this approach to constituting a smart city agenda and choosing verticals is that local authority departments that do not typically engage with digital technology innovations in their routine activities may be less likely to feature in smart initiatives, even if they face pressing problems.
Importantly, in most UK cities pursuing smart agenda, there is often a lack of a wellresourced dedicated smart city team within the local authority, and a lack of a dedicated funding pot to support their smart initiatives.A recent study in the United Kingdom uncovered that just over a quarter of local governments are very confident of the security of their organizations' digital networks and that only 15 per cent of local governments think that their digital infrastructure allows them to react effectively to change in working processes (Survey iGOV & Virgin Media Business, 2019).Also, in the United Kingdom, an estimated 876,000 jobs (representing a 30% reduction) have been lost in local government since June 2010 (GMB Congress, 2019, p. 15), resulting in "increased pressure on the remaining workers who are expected to do more with less" (GMB Congress, 2019).Owing to this, some leaders of smart programs feel compelled to steer their city's smart city agenda in directions that pragmatically allow for securing external funding and overcoming skills and labor deficits.The following insights from three council officers leading smart agendas in three different cities are instructive here: In terms of Smart Cities, we've delivered Smart City projects on a shoestring really.We don't have a Smart Cities team.Essentially, you're looking at him.So, I am [the] Smart Cities [team].-City A If there were 10 of me, 10 Smart Cities officers within the city council, then we would look at many different strands, many different opportunities.We can't cover all the different projects that there are.[Our smart city team] literally is just me.A lot of the other local authorities have Smart Cities teams.But I do it on my own -City B So while we defined the areas where we believe a smart city approach could support [improvements], we were limited in where we could get the funding.So we've done a lot of work in transport because we've been very successful in winning funding in that sector -City C Taken together the quotes above do not only reveal the extent of the resourcing challenges those leading smart city agendas across different cities face, they also illustrate the high level of agency these individual leaders have in choosing the direction of their city's smart city initiatives, including the selection of specific initiatives.This raises profound questions on democratic accountability in smart cities, and on how funding considerations are weighed against benefits to wider city stakeholders in the pursuit of specific smart initiatives (Ehwi et al., 2022).Furthermore, the fact that cities continue to market themselves as leading and innovative smart cities, despite significant operational challenges, shows the extent to which intercity competition has impacted transparency within the entrepreneurial governance system.

Vertical selection based on entrepreneurialism
The way in which particular verticals become the focus for urban authorities pursuing a smart city program also has a clear connection to entrepreneurial urbanism.Austerity has imposed cuts on local authorities, and so efforts to reduce spending or increase efficiency of existing services through smart initiatives might well be understood as a pragmatic response.However, there is also evidence of an entrepreneurial tone here, which is understood as necessary amidst the difficulties with managing a restricted budget: Resourcing [the Smart City programme] is difficult.We haven't got the skills or the number of people to manage lots of programmes.So, we rely on partnership working and bringing in likeminded people -Council Officer Public-private partnerships are, of course, a stalwart of entrepreneurial governance, but it is also notable that this council officer describes these partners as "likeminded".The implication of this appears to be that local authority interests are aligned with those of private investors.This would suggest an inclination toward supporting economic growth first and foremost, in keeping with broader trends in neoliberal urban governance.However, it is notable that some interview participants highlighted that supporting projects for which there was a clear business case was an essential exercise in fundraising for projects which did not have a clear economic justification: I think the parking [initiative] is sometimes seen as one where there is a very clear business case.And that can help pay for some of the others where perhaps the business case is less clear.-Smart Cities Consultant As such, while strategies which favor economic growthand thus result in the selection of smart city verticals most aligned to this goalindicate an entrepreneurial approach to urban governance, the pursuit of these entrepreneurial strategies is also part of an attempt by local authorities coping with restrictive budgets to make possible the delivery of projects which do not fulfill an economic growth-based criteria for selection (See Rossi, 2016 for variegation of smart city projects delivered through public-private partnerships in the austerity-hit Italian city of Turin).
Crucially, where exactly smart city governance sits within the local governance structure is of importance when it comes to shaping the intended outcomes of smart initiatives.Indeed, the positioning of a Smart City team within the economic development department of the local authority had a clear impact on the work of the Smart City team in Nottingham: My role sits in Economic Development.So anything I do, I'm constantly checked and pulled back, and it's like, "have you considered x, y, z local business in this?Please get them involved".-Council officer, Nottingham The involvement of private interests in urban governance is long established (see Harding et al., 2000).While this is not distinctive to smart city governance per se, there are clear implications for accountability, transparency, and democracy.The quote below highlights how in an attempt to be seen as apolitical, one case study city was taking care to ensure that businesses and commercial interests did not feel they were being "dictated to" by the (democratically elected) local political leadership: In the work that we've done [on the Smart City programme], we've engaged with our [city's elected] cabinet members … We know that without their endorsement and their leadership, a lot of this won't happen.So they will be instrumental in that.What role they'll play in the governance is yet to be determined … because what we don't want is [for the smart city to be] seen as a political issue.We want this to be a city issue, where city partners feel that they're not being dictated to by the city council, but they feel part of the family.So there might be some political overview.But I don't think at this stage we're intending to be driven by a political agenda, it'll be driven by a city agenda.-Council Officer, Birmingham Crucially, this council officer highlights that the local authority wants the smart city program to be perceived by (commercial) partners as driven by a city agenda, rather than a political one.Evidently, what counts as a "city agenda" here is one formed in conversation with business interests, rather than one directed by local politicians elected by city residents.Where business representatives direct the smart city agenda in this way, it makes sense that smart verticals which favor business interests are likely to be driven forward.Indeed, this further serves to highlight the importance of where decisionmaking sits within the overall governance framework in the development of smart city projects.

Vertical selection based on global and national imperatives
In addition to local pressures, there is evidence that broader policy imperatives (at global or national level) shape the selection of smart verticals.The ongoing climate crisis and the Covid-19 pandemic were commonly cited by interview participants as influences on the selection of smart initiatives.There was a general consensus that efforts to reduce and mitigate climate change needed to be considered in any smart city vertical because the impacts are not just limited to the environment but could also be seen in other areas such as healthcare and wellbeing, transportation, and the economy.For example, reflecting on how climate change became a key influence on the selection of verticals in the smart city agenda of Leeds, one interview participant observed: initially, climate wasn't on [our smart agenda] … If you've got a good environment, your health and wellbeing are likely to be improved.The work that you do on transport has an impact on the environment.So, for that reason, we didn't have climate or the environment as a priority area.But once we declared a climate emergency, I'd suggested that we should have climate emergency as one of our priority areas and really highlight that this is a big deal for us in Leeds -Council officer This quote illustrates that global events shape priorities for local investment.Notably, it appears that environmental considerations were already understood as important for Leeds' smart agenda given the effects of the environment on several priority areas, but until the city declared a climate emergency, this did not emerge as an explicit priority for its smart cities program.
There was also consensus that the Covid-19 pandemic has resulted in heightened calls for city leaders to embrace digitalization in all aspects of local service delivery.Specifically, it was argued that the pandemic has helped local authorities to appreciate how important it is to invest in smart infrastructure projects that furnish a real-time understanding of the city, in terms of, for example, changes in footfalls to city centers, levels of air pollution along major roads, and opportunities for local authorities to conduct some of their activities remotely.The quote below is illustrative of how the technology being used for one traffic-management initiative in Leeds was also found to be useful for monitoring changes to city center footfall: So highways are introducing some new technology and as far as they're concerned it's about traffic management.And yet that solution can help a different part of the Council to better understand how busy the city centre is, and certainly post-COVID, whether we're getting back to pre-covid levels of people that are going out into the city, and then they can access that data as well -Council officer It is also important to note that some of these global events feed directly into national policy agendas which in turn filter into city-level smart city initiatives.For example, in 2019, the United Kingdom government committed to achieving Net Zero carbon emissions by2050 (HM Government, 2021).Following this pledge, many local authorities have declared 2030 as the year to achieve their local net zero target and have established smart city verticals that align with this national ambition for a variety of reasons, including a possible opportunity to access grant funding linked to net-zero.Notably, officials from Nottingham stated that the city has a highly ambitious aim to achieve net zero by 2028, observing that: I was effectively given a blank canvas as to where to take Smart Cities in Nottingham, and … we're aiming to be carbon neutral by 2028.So we're the first of the major cities to be the most ambitious in terms of its timescale.So what I was asked to do was to connect the Smart Cities agenda to the carbon-neutral agendacouncil officer The quote above shows how the national policy agenda can impact local governance decision-making, and how those leading smart city initiatives steer the smart city agenda in line with broader policy shifts.Importantly, by linking the smart city program to a national (and international) policy agenda, the council in Nottingham was able to craft the city's image as ambitiousas well as presenting the city as competitive among other major citiesin a move characteristic of the use of smart city programs as a tool for urban boosterism (see Wiig, 2016).
Additionally, the United Kingdom government's decision to consolidate welfare benefit claims into the Universal Credit (UC) system which is accessible through an online portal was cited as influencing vertical selection.To avoid disadvantaging people without digital skills or access to the internet, many local authorities feature digital inclusion as a key vertical in their smart city agendas.The quote below from council officer in another case study city is illustrative of how national government policy directions do not only influence the selection of verticals at the local level but also alter the scope of activities within existing council departments.This was at a time when the government was rolling out Universal Credit.A lot of our tenants weren't online, didn't have the right skills, and the Universal Credit Scheme was online only because the government was shifting everything to online.So a lot of the people who needed to access Universal Credit couldn't get online and had never been online … This is where the Digital Inclusion team really started to expand -Councillor In short, this points to the nuanced ways in which the implementation of a smart city agenda and vertical selection are negotiated through complex interactions representing interest groups and policy changes at different spatial scales.This again raises the question of the extent to which smart cities meaningfully address local city problems.and how different drivers behind vertical selection interact to produce variable outcomes for urban residents.
Although this study identified four factors driving the selection of smart city verticals, it remains unclear whether all four are present in each of the case study cities.The fact that interview participants in a given city did not allude to a given factor does not necessarily imply that the factor left out is not applicable in the specific context.Similarly, where a number of factors are highlighted in one city does not mean that all are informing vertical selection to the same degree.We suspect that just as the expressions of neoliberal governance are known to vary across geography, modalities and pathways (Brenner et al., 2010), the balance of each of the four factors identified as driving vertical selection is likely to vary across cities based on distinct local contexts.
Further empirical research is needed to ascertain the degree to which each of the factors hold sway across different urban contexts.This notwithstanding, the four factors identified can be considered as a lens for thinking through the transformation that is happening in urban environments and in urban governance more broadly, as smart cities become commonplace.It can serve as framework for unpacking the underlying logics and drivers of smart city vertical selection that are often taken for granted or veiled beneath notions of efficiency and improved decision-making (see Batty, 2012;Kitchin, 2015).
Indeed, the logics which underpin the entrepreneurial governance observed across the case study cities can be usefully understood as representing various types of urban entrepreneurialism: To borrow from Phelps and Miao's (2020) classification of varieties of urban entrepreneurialism, justifications for smart initiatives which emphasize efficiency can be understood as "new urban managerialism", while efforts to capitalize upon the smart city label and associated image by boosting the investment potential of the city can be understood as "urban diplomacy".Tracing the logics of vertical selection also reveals something of the ways in which entrepreneurial approaches to urban governance can transcend growth agendas (see Lauermann, 2018), for instance, to enable the delivery of projects or developments for which there is not a clear "business case".Evidently, tracing the roots of vertical selection enables consideration of precisely what forms of entrepreneurialism are prevalent within particular cities, each with different historical contexts which shape governance approaches.
While popular narratives of smart cities paint them as technical, objective and apolitical, it is clear that this could not be further from reality (Kitchin, 2015).Drawing attention to the ways in which certain areas of focus for smart initiatives emerge, and to the variation in smart programs which different drivers for vertical selection produce across different cities, highlights that the processes of decision-making in smart city governance are just as political and complex as in other well-established areas of urban governance.

Figure 1 .
Figure 1.Number of different Smart City Demonstrators in UK Cities.Source: Compiled from the Future Cities Catapult's Smart City Demonstrators: A Global Review Report (2016).

Funding
The author(s) disclose receipt of the following financial support for the research, authorship, and/ or publication of this article: This research forms part of the Centre for Digital Built Britain's (CDBB) work at the University of Cambridge within the Construction Innovation Hub, which brings together world-class expertise from the Manufacturing Technology Centre (MTC), BRE and CDBB to transform the United Kingdom construction sector.The Construction Innovation Hub is funded by UK Research and Innovation through the Industrial Strategy Fund under Grant Number RG96631.

Table 1 .
Brief information about the selected case study cities.
Note: Data on city features A, B, D & E were sourced from 2022 NOMIS Local authority Profile.NOMIS is a service provided by Office for National Statistics (ONS), the UK's largest independent producer of official statistics.It publishes statistics related to population, society and the labour market at national, regional and local levels.These include data from current and previous censuses (URL https://www.nomisweb.co.uk/)CityInformation Feature C was sourced from GOV.UK (2012)