Policy failure mobilities

Abstract

This paper concerns the movement of negative lessons and worst practice in public policy. It focuses on a relatively new branch of scholarship – policy mobilities – which explores the global movement of policies. Within policy mobilities research there is concern about an empirical bias towards successful policies: there has been insufficient attention to whether policy failures might also be mobile. Ideas and concepts about policy failure from political science, economic geography, and science and technology studies are used to illuminate what is missing from policy mobilities scholarship, why it might be important, and to offer some ways forward.

Keywords

lesson-drawing policy failure policy learning policy mobilities worst practice

I Introduction

My interest in cases of policy worst practice and their mobilities was sparked by research on learning from smart grid experiments in Australia. Interviewees repeatedly told me they had been motivated to act in a particular way in order to avoid ‘doing a Victoria’ – referring to a mandatory smart metering program implemented in the State of Victoria in the period 2009–13, called the Advanced Metering Infrastructure (AMI) Program. The AMI Program encountered problems with high costs, smaller than expected benefits, negative media coverage and household resistance (VAGO, 2015), and it was evident that this negative policy example was having an influence on decision-making across Australia, beyond the State of Victoria (see Lovell, 2017). For me, the case raised questions about the geographies of worst practice policies, including how far negative policy lessons travel, and whether their mobility and effect is substantially different to the best practice, successful policy cases that currently represent the majority of empirical research within scholarship on policy mobilities.

The main aim of this paper is thus to highlight, and attempt to rectify, a bias in scholarship on the geographies of policy towards ‘best practice’ empirical cases, with policy mobilities scholarship as its primary focus. It is suggested that deeper attention to three disciplines or sub-disciplines that policy mobilities scholarship has emerged from – namely, political science, economic geography and science and technology studies (STS) – that are for the most part attentive to both positive and negative forms of learning, helps provide a more balanced perspective and approach. The objective of the paper is thus to build on the conceptual influences already present in policy mobilities scholarship in order to develop a clearer understanding of the movement of negative policy cases or ‘worst practice’. These conceptual influences include: political science research on typologies of policy failure and the politics of framing something as either a success or failure; economic geography theories about proximity, place and learning and the distinction between tacit and codified knowledge; and STS research on experimentation and the inherent fragility of sociotechnical networks.

It is acknowledged that the notion of policy mobilities also, of course, has foundations in mobilities theory (Adey et al., 2014; Cresswell, 2012; Urry, 2007). Mobilities theory positions travelling and movement as the prime objective of study: ‘it focuses on, and holds centre stage, a fundamental geographical fact of life – moving’ (Cresswell, 2010: 551). Topics of study are varied and include transport, the mobility of ideas and things, gendered mobilities and global mobilities of labour and capital (see Adey et al., 2014). However, mobilities theory does not have a central focus on learning, and for this reason it is not considered in detail in this paper.

The paper is structured as follows. First, scholarship on policy mobilities is outlined. Its main arguments are summarized, and the extent to which it has conceptually and empirically engaged with policy failure is examined. Second, scholarship that attends more closely to policy failure from across the disciplines of political science, economic geography and STS is reviewed, and the insights from these explorations of policy failure are considered. In conclusion, the main insights from the cross-disciplinary review are summarized, their significance for human geography research on policy mobilities is outlined, and potential future research avenues are suggested.

II Policy mobilities: An introduction

The concept of policy mobilities has been developed principally by urban studies and human geography scholars in an effort to capture empirical observations about the contemporary international movement of policy, drawing on a range of ideas from outwith political science (the core discipline traditionally dominating research in this field) including mobilities, materiality and processes of economic development and innovation (McCann, 2011; Peck, 2011; Ward, 2006; Baker et al., 2016). Labelled ‘a new wave’ of policy transfer research (Hudson and Bo-Yung, 2014: 495), a number of novel ideas and concepts have been introduced by policy mobility scholars. First, the observation that policy ideas, knowledge and programs are circulating internationally with greater ease and speed – a phenomenon linked to wider processes of globalization and neoliberalism – and termed ‘fast policy’. The notion of ‘fast policy’ conveys the increasing time pressure under which policy decisions are having to be made, described by Peck (2011: 773) as follows:

Today’s ‘fast-policy’ regimes are characterized by the pragmatic borrowing of ‘policies that work’, by compressed reform horizons, by iterative constructions of best practice, by enlarged roles for intermediaries as ‘pushers’ of policy routines and technologies.

Note here too the specific focus on ‘best practice’. Second, policy mobility scholars have documented the ways in which policy is increasingly being made beyond formal, official government channels and venues by a range of non-state actors or intermediaries, such as multi-national companies and consultants (McCann, 2011; McCann and Ward, 2012; Prince, 2012; Larner and Laurie, 2010). This is perhaps its core distinction from political science research on policy transfer and diffusion, which first emerged in the 1960s (see, for example, Walker, 1969), wherein policy transfer is defined quite generally as ‘a process in which knowledge about policies, administrative arrangements, institutions and ideas in one political setting (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political setting’ (Dolowitz and Marsh, 2000: 5). But, despite the use of the open-ended term ‘political setting’, the majority of political science scholarship has nonetheless concentrated on government-to-government transfer: the movement of policy between a small number of countries, most typically through instances of bilateral transfer, as well as policy movement between regions within nation-states (Bennett and Howlett, 1992; Dolowitz and Marsh, 2000; Duncan, 2009; Mossberger and Wolman, 2003; Walker, 1969, although see Stone, 2004, for an exception ). In contrast, and drawing on its wider disciplinary heritage, policy mobilities scholarship has analysed the role of a much broader mix of organizations, examining how private sector actors in particular initiate and facilitate the international movement of policy ideas, knowledge and programs (Pow, 2014; González, 2011), and with a focus on urban centres and large cities rather than nation-states (McCann, 2011; Clarke, 2012). Prince (2012), for instance, analyses the role of international cultural consultants in the dissemination of new policy ideas and practices, and, in an early contribution to the field of policy mobilities, Ward (2006) provides detailed study of the movement of the policy of business innovation districts between the US and the UK. Scholars such as Wood (2015) and Didier et al. (2012) have broadened the empirical reach of study to the Global South, with Wood (2015), for example, examining the rich politics of bus rapid transit policy mobility from the city of Bogota in South America to South African cities, and Didier et al. (2012) analysing the transfer and adaptation of ‘city improvement districts’ between Johannesburg and Cape Town.

Third, policy mobilities scholarship is attentive to the materiality of policy, that is, the objects and things that constitute policies, exploring how they are an integral part of the policy process (Peck and Theodore, 2015 ). Policy mobility scholars are particularly interested in how policies change or mutate as they move, as McCann (2011: 111) explains:

Policies, models, and ideas are not moved around like gifts at a birthday party or like jars on shelves, where the mobilization does not change the character and content of the mobilized objects.

Fourth, there is a strong focus on the heterogeneous networks comprising people and technologies that constitute policy, termed ‘assemblages’ – a concept originally developed within STS (also variously referred to as agencement or actor-networks; see, for example, Hardie and Mackenzie, 2007; Law and Hassard, 1999) – and defined within policy mobilities scholarship as ‘a purposive gathering of people, institutional capacities, expertise, models, techniques and technologies, political sustenance…from local sources and, crucially, from elsewhere’ (McCann, 2011: 144). As McCann alludes to in his reference to assemblages constituted by ‘local sources and…from elsewhere’, an integral feature of policy mobilities scholarship is the study of relational geographies, i.e. relations between people and things which are not necessarily characterized by topographical proximity. In this way the use of assemblage within policy mobilities research builds on its economic geography heritage – it is in keeping with theorizations of change and innovation within this sub-discipline, for example Allen and Cochrane’s (2007: 1161) definition of a regional assemblage approach as ‘a relational view of the region based upon an assemblage of political actors, some public, some private, where elements of central and local government are “lodged” within the region, not acting above or below it’. The use of assemblage within policy mobilities research brings to the fore these relational geographies – the linking of different places and spaces – more than the original STS meaning, which has the combining of the social and the technical as its main focus. Drawing inspiration from Deleuze and Guattari’s (2004) fluid way of thinking about the social world as comprising heterogeneous networks of things, objects and ideas, STS scholars such as Callon (2007) and Hardie and Mackenzie (2007) have used the term assemblage or agencement in a slightly different way to describe and explain processes of sociotechnical innovation, with a strong focus on distributed agency, a point returned to below.

In summary, policy mobility scholars have explored the broader context within which policies emerge and move, with close consideration of the dynamic political, economic and sociotechnical relations of a wide range of actors (human and non-human, public and private), and with a particular focus on cities and globalization, thereby broadening the lens of longstanding political science policy transfer scholarship, both conceptually and empirically.

1 The hidden geographies of policy mobility: Theorizing policy failure

Since the emergence of the concept of policy mobilities in the mid-2000s there has been growing recognition of an empirical gap, namely a shortage of research on the movement of policy that considers negative lesson-drawing, or learning from mistakes and policy failure (McCann and Ward, 2015; Webber, 2015; Jacobs, 2012). The remainder of the paper concentrates on this area of neglect – the underbelly of policy transfer and mobility. For, despite the upswell of activity and new engagement in the geographies of policy, there remains a preponderance of case studies of successful best practice policies. In other words, policy mobilities research is overwhelmingly about policies that do work and are ‘present’ – publicly promoted and discussed as successes – ranging from urban regeneration in Bilbao and Barcelona (González, 2011), to sustainability in Vancouver (McCann, 2008), and business improvement districts in New York (Ward, 2006). Scholarship is therefore in effect missing a large part of the empirical picture. Instances of immobility or non-transfer because of policy failure, as well as explorations of how and why negative policy lessons circulate, are issues that have been empirically neglected (albeit with some exceptions, discussed further below; see Lovell, 2017; Müller, 2015; Webber, 2015).

Conceptually, however, there has been increasing recognition of the need to be more attentive to policy failures, for example in calls for greater recognition of the role of power and politics in shaping what is seen (and what remains hidden) in policy development, as Peck (2011: 791, emphasis added) explains:

the field of policy transfer…is saturated by power relations. These intensely contested and deeply constitutive contexts…shape what is seen, and what counts, in terms of policy innovations, preferred models, and best practices.

Jacobs (2012: 419) more specifically calls for approaches that consider ‘Sites of failure, absence and mutation [as] significant empirical instances of differentiation’, and draws attention to the methodological bias of policy mobility scholarship which is ‘fixated on policy presences, following what has already arrived and formed’ (2012: 418). Clarke (2012), taking a distinctive historical approach, also makes a considered and thoughtful call for scholarship on policy mobility to be attentive to the wider dynamics of policy change and decision-making, including the policy ideas and programs which remain immobile. Drawing on the work of the urban historian Saunier, Clarke’s criticism of policy mobilities scholarship is, like Jacobs, primarily methodological. He argues that the ‘flow’ methodology advocated by mobility scholars both encourages attention towards successful policies as well as restricts exploration of the wider context surrounding these flows, whereas Saunier’s ‘[historical] starting point brings into focus the full breadth of attempts to mobilize urban policy, some of which succeed while others fail. This breadth of view is lacking in studies of contemporary urban policy mobility’ (Clarke, 2012: 39, emphasis added). More recently, McCann and Ward (2015: 1) have identified policy success and failure as well as policy mobility and immobility as two out of the three ‘dualisms’ or tensions within current policy mobilities scholarship, noting how:

Neither success nor failure is absolute. One does not make sense without the other. Rather, success and failure are relationally constituted in politics and in policy-making.…Much of the work in the urban policy mobilities approaches has, almost by definition, emphasized those policies that appear to be ‘mobile’, where there is evidence of the policy being moved from one location to another.…The ‘other’, so to speak, in the literature is the group of policies that do not appear to have travelled, policies that appear to exist in just one location. (2015: 2, emphasis added)

Such comments build on earlier contributions by McCann (2008: 16, emphasis added), who raised a number of important questions about immobile policies:

it is worth tempering talk of policy mobilities with investigation into policy immobilities…the danger of privileging and fetishizing mobility needs to be avoided…a number of questions follow: which policies are not mobilized, why, and who is impacted positively or negatively by this immobility?; how do…subjectivities influence which policies are, or are not mobilized?…can ‘nonmainstream’ policies take advantage of global circuits of policy knowledge?

But the implication here in identifying ‘nonmainstream’ policies appears to be that immobile policies are successful but overlooked, rather than failures. McCann’s comments thus draw our attention to a degree of conceptual ambiguity about the relationship between immobility and failure within policy mobility scholarship. It seems probable that policies that have not worked well are mobilized and travel in different ways to successful best practice case studies, but is it the case that they do not move at all? There is an overall lack of clarity about what constitutes immobile policies: whether it be successful but unrecognized policies, or failed ones, or a mix of both.

Another uncertain issue is what is mobile in cases of policy failure. As noted, the substance of policies has been a core interest of policy mobility scholars, but to date – in keeping with the dominant trend in policy mobility scholarship – attention has been largely directed at the making and holding together of policy successes (Geddie, 2015; Temenos and McCann, 2012). This emphasis on the development of stable policy networks is, however, somewhat at odds with a rich heritage of STS scholarship deploying the idea of assemblage¹ to explore the fragility, breakdown and failure of networks: from Callon’s (1986b) classic case of scallop fishing in France, to examples in agriculture (Higgins and Kitto, 2004), housing (Lovell and Smith, 2010) and medicine (Singleton and Michael, 1993). It is argued below that casting the conceptual net wider to embrace STS ideas about the breakdown of networks helps better recognize and conceptualize the movement of policy failures.

Within the policy mobilities field there are, however, a small number of studies that do empirically consider policy failure. A case study of smart metering policy in Australia demonstrates the national mobility of a policy failure, but with limited international movement (Lovell, 2017). Webber’s (2015) research on the mobility of World Bank climate adaptation initiatives between Kiribati and the Solomon Islands shows how, despite World Bank initiatives in Kiribati not working well (and indeed the second program ‘KAP-II’ being judged unsatisfactory by the World Bank’s own mid-term review process), the Kiribati program was nonetheless promoted as a success in several World Bank reports, and used as a model elsewhere. Webber stresses the importance of ‘Attending to the differences between what is mobilized in rhetoric and in practice’ (2015: 36), hence drawing an important distinction between the discourses used by actors involved in the movement of policies, and the practices of implementation. This point is echoed by Müller (2015) in his analysis of environmental policy mobility using the case of the 2014 Olympic Games in Sochi, Russia. Müller identified in his research on Sochi the problematic and incomplete implementation of practices and approaches used in the previous Vancouver Winter Olympics, noting how ‘where there is transportation, there is also immobility: certain people and things stay put, they escape the attempt to move them or move only partially’ (2015: 195).

Whilst the cases discussed by Webber and Müller are vital contributions, they are not, however, at their core about the movement of failed policies: Webber’s climate adaptation case is about the movement of a policy that failed in parts, but nonetheless was discursively framed and travelled as a success. Müller’s case is about the import of a policy success (from the Winter Olympics in Vancouver) that subsequently mostly failed. These empirical cases raise questions about the different types of policy failure – an issue that has been studied within political science (see below) – but which has not yet been clearly addressed within policy mobilities research. Political scientists such as Marsh and McConnell (2010), for instance, have classified policy failure as either process, programmatic or political failure, thereby taking the stages of the policy process (decision-making, implementation, post-implementation) as a springboard for classification. As is discussed below, whilst political scientists tend to be focused on temporal processes of policy-making within the nation-state, policy mobilities research is instead most interested in the geographies of policies. Any classification of policy failures within policy mobilities research thus has a quite different conceptual starting point comprising the geographies of failure and their materialities, i.e. what moves and what does not with regard to failed policies, and how immobility contributes to, or helps avoid, policy failure. Webber, in her analysis of a World Bank climate change adaptation project in Kiribati, encapsulates some of these geographies of policy failure in her rich description of the ‘multiple failures and stoppages…[including]…failure in project outcomes, failure to mobilize, failure to implement in replication sites, and, most importantly, failure that becomes success through iterative extraction and interpretive processes for extra-local learning’ (2015: 29).

Müller too talks about how ‘certain people and things stay put’ with regard to the Olympic Games and how this partial immobility is what contributes to policy failure. In contrast, in the case of smart meters analysed by Lovell (2017), what is mobilized are particular discourses about policy failure, whilst information and reports regarding the policy successes (e.g. the high implementation rate of the new meters, enabling excellent ‘visibility’ of network) have not circulated, nor has much detail around what precisely it was that did not work well, as Lovell (2017: 325) describes: ‘The AMI [Advanced Metering Infrastructure program] has predominately travelled discursively – as a story of policy failure’.

These cases draw our attention to the varied materiality and mobilities of policy failures. The materiality and material encounters of failed policies include the site visits, reports and study tours that typically populate policy mobilities research on successes and best practice policies (see Müller, 2015, and Webber, 2015), but so too are there absences: Lovell (2017), for example, describes the absence of discussion of the Victorian smart metering program in reports produced by international smart metering organizations. The case of smart metering is also interesting in the attempts made by some organizations to ascribe and isolate the policy failure to the prototype technology involved – the smart meter – which was in its infancy when the program was first agreed to, as the head of smart metering at a large international smart metering company is quoted as saying: ‘the Victorian problems emanated from decisions taken almost 10 years ago, and since that time the technology has “leapfrogged”’ (cited in Lovell, 2017: 322).

Bringing the themes of spatiality and materiality together, Robinson’s (2015) observations about the topographical and topological spatialities of policy mobilities are instructive here in considering the ‘imaginations’ of policy, beyond visible policy reports, buildings and other material forms of policies, as she describes: ‘Arriving at policies involves far more than assembling discrete materialized entities, ideas or objects which we can trace as they move from there to here. Complex, topological spatial imaginations are needed to interpret the mixing and folding of here and “multiple elsewhere” (Mbembe and Nuttall, 2004) into distinctive local policies’ (Robinson, 2015: 833, emphasis added). The topological spatialities of policies are arguably especially important with regard to policy failures, through the spread of stories, images, and fears of failure, which may circulate quite independently of the material substance and detail of what took place with any particular policy.

Having introduced the concept of policy mobility, and discussed growing concerns around its empirical bias towards the study of successful policies, as well as work to date on policy failure mobilities, attention now turns to consider a potential solution, namely engaging in more depth with political science, STS and economic geography scholarship on learning, and in particular concentrating on insights from these three fields about failure and negative learning and lesson-drawing.

III Scholarship that does attend more closely to policy failure

Political science, STS and economic geography are the three broad areas of scholarship that the concept of policy mobilities has emerged from, and it is these conceptual origins of policy mobilities that are concentrated on here, because there are already influences, present within policy mobilities scholarship, that can be built upon. Within political science and economic geography there is, in a similar way to policy mobilities scholarship, an emphasis on purposeful, directed, lesson-drawing with a bias towards best practice policies. However, these disciplines, along with STS, have also been attentive in different ways to negative learning and failure, and it is this specific area of enquiry that is reviewed below, with a view to providing a range of insights for policy failure mobilities.

1 Political science

As noted, geographers and urban studies scholars have drawn on the political science concepts of policy transfer and diffusion in order to develop new ideas about policy mobilities (Dolowitz and Marsh, 1996; Walker, 1969). Political science is thus one of the core conceptual foundation stones for policy mobilities, and it is to this branch of scholarship that attention is first directed in searching for insights about negative learning and lesson-drawing. For there is existing political science research on this topic which to date has arguably been rather overlooked by policy mobilities scholars, that considers negative lesson-drawing and learning from policy failure (May, 1992; Marsh and McConnell, 2010; Howlett, 2012; Robertson, 1991; Bovens and t’Hart, 1996; Illical and Harrison, 2007). Bennett and Howlett (1992), for example, identify mixed interpretations (positive and negative) of the same policy, explaining how:

Learning…can be both positive and negative. That is, learning is both about what to do, and about what not to do, so the same program can act as a model or exemplar for one country, and exactly the reverse for another. (1992: 291, emphasis added)

Bennett and Howlett thus acknowledge the potential for fluidity in meaning in what constitutes a policy success or failure. Such an approach considers the politics of framing a policy as one or the other, at the same time drawing our attention to the geographical nuances of such processes, wherein key decision-makers in one country or locality can interpret a policy implemented elsewhere as a success, and another frame the same policy as a failure. This dualism or tension between policy success and failure has, as noted above, recently been identified by policy mobilities scholars as an issue ripe for further research (McCann and Ward, 2015). Howlett also acknowledges the degree of political and discursive work involved in classifying a policy as either a success or failure, noting how ‘designations of policy success and failure are semantic or “discursive” tools themselves used in public debate and policy-making processes in order to seek political, partisan, and often electoral advantage’ (2012: 547).

Marsh and McConnell (2010) have done valuable work in distinguishing between different types of policy failure or success: process, programmatic and political, with process referring to the early stages of policy-making (passing legislation, etc.), programmatic to the operational stage of the policy, and political to its popularity during and after implementation, and the repercussions for the government (Marsh and McConnell, 2010: 571). This is a helpful typology, but there remains – as they recognize – a difficulty in clearly distinguishing between success and failure, noting that: ‘“success” will always be contested to some degree’ (2010: 575).

In addition to these conceptual contributions, political science scholarship also encompasses a number of valuable empirical studies of negative lesson-drawing and policy transfer. Illical and Harrison (2007) provide a rare, detailed long-term study of negative learning about the transfer of endangered species policy between the US and Canada. The US implemented the Endangered Species Act in 1973; Canada learnt from some of the things that did not go well and, a long time later – in 2002 – implemented its own version of the policy. A core finding of this study is the wider context or ‘fit’ that shaped Canadian policy, as the authors explain:

The case of endangered species policy in the US and Canada thus offers not only a clear example of the impact of negative lesson-drawing, but also suggests the importance of the interaction of lesson-drawing, interests, and institutions.…the negative lessons of the US Endangered Species Act were readily incorporated because they ‘fit’ with the institutional setting and prevailing balance of interests. (Illical and Harrison, 2007: 390)

This finding echoes early analysis by Robertson (1991) in his research on decision-making across several policy issue areas in the US, including mandatory plant closures, labour markets, and income support. Robertson provides a detailed account of how positive and negative lesson-drawing from elsewhere is more or less likely to occur at different stages of the policy process: with positive lesson-drawing tending to take place at the early agenda-setting stage, and negative lesson-drawing during intense decision-making on implementation. This insight is an important one for policy mobility scholarship, where issues about the temporality of policy change processes, and how this intersects with the spatiality of policy, have yet to be considered in detail (see Clarke, 2012, for a notable exception). More broadly, in his analysis Robertson makes a strong argument for attending to the politics of policy transfer, noting how ‘Policy lessons from abroad often are put forward as politically neutral truths. Beneath this superficial impartiality, political adversaries…are using such lessons as political weapons’ (Robertson, 1991: 55). Such ideas resonate strongly with policy mobilities research, which is attentive to the inherent political geographies of policy decision-making (see, for example, Robinson, 2015).

So, in summary, a number of political science scholars have been attentive to negative learning, and their findings are instructive for policy mobilities scholarship. For instance, Robertson’s and Illical and Harrison’s empirical studies strongly suggest that policy failures have agency beyond the particular locality in which they were first implemented. These contributions also point to potentially interesting temporal dynamics around the movement of policy failures, including time lags, and moments within the policy process when the circulation of policy failures is more likely. However, a key limitation of political science studies is that the overall emphasis of these approaches is about learning within government, rather than whether or how these policy failures might be constituted and learnt from by a broader range of actors. For this reason attention now turns to the discipline of economic geography, with its empirical focus on corporate and other non-state actors.

2 Economic geography

One of the most interesting and important insights within economic geography about learning with regard to this paper’s remit is its close attention to different types of knowledge – tacit and codified – and their respective mobilities. The categorization of knowledge as either tacit or codified was first developed by Polanyi (1966) (although he termed codified knowledge ‘explicit knowledge’). The basis for the idea of tacit knowledge is described by Polanyi as ‘we know more than we can tell’ (1966: 4), and is defined more extensively as knowledge which is ‘not fully reducible to a clearly articulated set of axioms, rules, algorithms, and statements’ (Gertler, 2003: 77) and as ‘learned behaviour or “know-how” [which] is intrinsically centred on the individual’ (Bunnell and Coe, 2001: 581). In contrast, codified knowledge ‘involves know-how that is transmittable in formal, systematic language and does not require direct experience…it can be transferred in such formats as a blueprint or operating manual’ (Howells, 2002: 872).

The links between these two types of knowledge and geography – proximity, place, scale – are immediately apparent, and this has been the focus of much economic geography scholarship regarding knowledge and learning (Bunnell and Coe, 2001; Howells, 2002; MacKinnon et al., 2002; Morgan, 2004; Rutten, 2016). The core idea is that codified knowledge, which can be translated, packaged and standardized, is more easily able to travel, whereas more complex and subtle tacit knowledge tends to be fostered and shared locally, through face-to-face contact, as McKinnon et al. (2002: 301) summarize:

While codified knowledge can be easily traded or communicated through markets and hierarchies, and can in principle become ubiquitously available, tacit knowledge is much ‘stickier’, being embedded in production practices and the ‘know-how’ of particular firms and workers.

And Gertler (2003: 79) further explains:

tacit knowledge, because it defies easy articulation and is best acquired experientially, is difficult to exchange over long distances…its context-specific nature makes it spatially sticky, since two parties can only exchange such knowledge effectively if they share a common social context, and thus important elements of this social context are defined locally.

A related concept to tacit knowledge, used by Bathelt et al. (2004: 38) to capture learning which is informal and spontaneous, is ‘local buzz’, defined as:

the information and communication ecology created by face-to-face contacts, co-presence and co-location of people and firms within the same industry and place or region. This buzz consists of…intended and unanticipated learning processes in organized and accidental meetings…Actors continuously contribute to and benefit from the diffusion of information, gossip and news by just ‘being there’. (Gertler, 1995)

There is a wealth of economic geography scholarship on learning regions, operating at a subnational scale in cities and other localities, which rests on the idea that tacit knowledge sharing and learning is fostered within places with a ‘common social context’, variously termed industrial districts, clusters, innovative milieu, neo-Marshallian nodes, technology districts, and technopoles (Bunnell and Coe, 2001: 575).

Whilst much of this scholarship focuses on successes, there is research that considers the sharing of knowledge and learning about failures. The distinction between tacit and codified knowledge is central to this analysis, for it is argued that tacit knowledge includes discussion of failures, but not codified knowledge. Bathelt et al. (2004), for example, describe how the tacit knowledge-sharing inherent in ‘local buzz’ includes sharing of failures and mistakes, whereas codified, globally-circulated knowledge sharing (flowing through ‘global pipelines’) is largely about successes:

In contrast with the information flows in global pipelines, the local buzz spreads information of both the successes and failures of other actors and their projects. To go beyond the regional cluster and engage in global pipelines is, to some extent, more risky because information flows about other firms are biased towards successful endeavours at this level and tend to overlook the not-so-successful. It is especially the outstanding successes which make their way through global communication channels. (Bathelt et al., 2004: 51, emphasis added)

In other words, Balhelt et al. argue that learning from things that go wrong is much more likely to occur within a particular locality, with actors that are topographically proximate – through informal and spontaneous face-to-face contact involving tacit knowledge sharing – and that the international mobility of lessons is restricted to successes. This is a crucial distinction, which goes some way to situate and explain the overall finding of policy mobility scholars, with their focus on the global flows of policy ideas and programs, that what is mobile at this scale is mostly best practice and successful policies. Research on the movement of best practice sustainability policies concurs with this economic geography finding, as Bulkeley (2006: 1041) explains:

The creation and use of best practice as a means of reward and recognition for particular initiatives, individuals, and places means that only ‘good news’ stories are disseminated, and that the (sometimes) murky details of how practices were put into place are obscured.

Economic geographers have a long-standing interest in relational geographies – an approach also adopted by policy mobilities research, as noted – that seeks to blur the boundaries between what is global and local, and problematizes the meaning of proximity, as Bunnell and Coe (2001: 570) explain:

Our key argument…is that there is a need for a qualitative shift away from work which focuses on particular scales as the locus for understanding innovation, towards that which gives more credence to relationships operating between and across different scales.

In this way a relational geography approach challenges traditional definitions of ‘local’ or ‘proximate’, with such terms expanded so they are no longer defined as solely about topographical proximity, or physical nearness (Bathelt, 2006). Such thinking has likewise been a strong influence in the development of the concept of policy mobilities – centred as it is on highly mobile globally distributed networks of policy actors, who nonetheless have close networks, shared knowledge, culture and work practices (see, for example, Prince, 2014; Peck, 2011). A relational approach posits that tacit knowledge development and exchange can be rethought as not necessarily taking place in a fixed locality, amongst physically proximate actors. There is thus a blurring of spatial boundaries that is important here in thinking about knowledge development regarding policy failures. Some scholars such as Bathelt et al. (2004) and Morgan (2004) argue that topographically proximate learning (‘local buzz’) retains particular qualities, including trust, social depth and chance face-to-face encounters, which boost the potential for learning from failures. But others contend that tacit knowledge is as easily developed across relational networks of actors who are not necessarily in the same place (Rutten, 2016). This is a large, complex debate, and not one that it is possible to summarize in full here (see Jones, 2013; Fløysand and Jakobsen, 2011). The key insight from economic geography scholarship for this paper lies in its distinction between tacit and codified knowledge, and there is judged to be considerable scope to apply this distinction to policy mobilities scholarship, investigating, for example, the forms of knowledge sharing – whether tacit or codified – taking place amongst global networks of policy actors and, for instance, how this might vary from sector to sector, and/or over time.

3 Science and technology studies (STS)

STS scholarship has an inherently more open view of learning and its potential outcomes, i.e. it is equally attentive to both positive and negative learning. In this way it is distinctive from political science and economic geography, which, as noted, have – notwithstanding the scholarship on failure analysed above – an overarching focus on learning from positive best practice case studies. The attentiveness of STS to failure most likely stems from its empirical origins studying innovation and experimentation in science laboratories, where experimental failure is viewed as a commonplace and integral feature of the scientific process (Shapin and Schaffer, 1985; Latour, 1987b). Key scholars such as Bruno Latour spent time observing scientists in laboratories and later in the field – providing detailed ethnographies of scientists and their day-to-day work practices and forms of knowledge development, sharing and learning, as described, for example, in Laboratory Life (Latour and Woolgar, 2013) and Science in Action (Latour, 1987b). Thus a core interest of STS scholars is the ways in which things don’t work, their inherent fragility, and the propensity of new ideas, experiments and programs to unravel. Actor-network theory (ANT) is one way in which this interest has developed (Law and Hassard, 1999; Callon, 1986a). Within ANT, the process of translation is the route by which actor-networks are conceptualized as coming into being. Translation involves painstaking work in bringing different human or non-human actors together (termed ‘enrolment’). Callon, for example, describes the process of translation in his classic case study of scallop fishing in the Bay of St Brieuc, France, which ultimately is a case of failure, despite the hard work by the fisheries researchers and other actors to make the new scallop larvae generation methods successful: in the end the scallop larvae fail to thrive in new specially designed collector units, and they are harvested too early by the fishermen (Callon, 1986). As Callon describes, ‘translation is a process, never a completed accomplishment, and it may (as in the empirical case considered) fail’ (1986: 196). Similar ideas about the tendency of heterogeneous or sociotechnical networks – comprising human and non-human entities – to disintegrate or fail have been applied to a whole range of empirical areas, from agriculture (Higgins and Kitto, 2004) to utility infrastructures (Guy et al., 2001). For example, Graham and Marvin in Splintering Urbanism describe how ‘Infrastructure networks, are, in short, precarious achievements’ (2001: 182).

It is suggested that there is scope to analyse how STS insights about failure and things not working can be better integrated into policy mobilities scholarship, through considering closely the fragility of policy networks, and the ways in which they break down (see, for example, Lovell, 2017). As noted above, the use of assemblage within policy mobilities scholarship has not significantly carried forward the core notions of distributed agency, fragility and failure which populate STS scholarship. Assemblage, as defined by policy mobilities scholars to date, has instead been more about the (largely successful) bringing together of disparate things and people from a range of localities (Cochrane and Ward, 2012).

There are two ways in which STS scholars conceptualize fragility which are especially relevant to the analysis of policy failure. First, in studies of the public demonstration of successful science experiments (see, for example, Shapin and Schaffer, 1985) and the precarious way in which these experiments were promoted and held to be successes, through performance. Experiments in the 17th century involved taking the laboratory quite literally to the public: key scientists such as Boyle and Hobbes set up experiments in public venues and used them to convince others of the experiments’ validity and rigour (Shapin and Schaffer, 1985). STS scholars view these public demonstrations of science as both highly politicized, and uncertain of success (Callon, 1986; Gieryn, 2006). Such a viewpoint provides useful reflection for policy mobilities scholarship, for it suggests that overt policy successes rely on political ‘work’ to underpin them, including overlooking or ignoring things that haven’t succeeded. It draws our attention to the fragility of successful policies, and the ways in which they are performed to relevant publics.

Such themes have been usefully developed in work on urban experiments and living labs (Leminen et al., 2012; Evans and Karvonen, 2014; Karvonen and Van Heur, 2014; McFarlane, 2011), and indeed are also acknowledged within policy mobilities research (McCann and Ward, 2015), but they could be developed more fully by empirically attending to instances of policy failure.

Second, the STS concept of immutable mobiles is important because it runs counter to the policy mobilities idea of mutation; of policies changing as they circulate and diffuse from place-to-place. The concept of an immutable mobile was first developed by Law (1987) and Latour (1987a), amongst others. It describes the phenomenon by which successful networks propagate through particular objects – immutable mobiles – defined as ‘that which moves through…space while holding its shape’ (Law and Mol, 2001: 619 ). Immutable mobiles remain stable even in different contexts – allowing otherwise fragile assemblages to maintain their coherence and be mobilized – thereby facilitating ‘action at a distance’ (Latour, 1987; Law, 1987). The concept arose from Law’s analysis of Portuguese imperial expansion in the 15th and 16th centuries, investigating how ‘long-distance social control’ came about, seen as essential to the imperial aims and capabilities of Portugal. The main theoretical claim is ‘that the undistorted communication necessary for long-distance control depends upon the generation of a structure of heterogeneous elements containing envoys which are mobile, durable, forceful and able to return’ (Law, 1986: 257), and thus there is a contrast here to the mutability of policies described by policy mobility scholars (Peck, 2011). But there is perhaps value in considering that those involved in policy decision-making nonetheless hold an ideal or ambition of a policy ‘immutable mobile’, i.e. a belief that policies can be transported from place to place without changing or unravelling, which in turn influences policy decision-making and practice. For example, this could be an alternative way of conceptualizing the policy transfer failure described by Müller (2015) in his analysis of the Olympic Games, where the ambition of moving best practice from one Olympic Games location to another does not in reality eventuate as originally conceived: the immutable mobile fails to hold together.

IV Summary and conclusions

This paper has sought to explore the issue of policy failure mobilities through investigating how policy failure is conceptualized by three key foundational areas of scholarship for policy mobilities, namely political science, economic geography, and STS. All three disciplines or sub-disciplines have attended to policy failure, albeit in different ways. Here in conclusion I briefly summarize their main insights, and then move to consider the implications for the study of policy failure mobilities. A key conclusion is that analysis of policy failures does not only necessitate a shift in methodology, as has hitherto been called for (Jacobs, 2012), but also requires a more profound change in the conceptualization of policy mobilities. Second, that political science, economic geography and STS offer different conceptual paths – that mostly diverge, rather than converge – and a suggestion is made about what appears to be the most promising way forward, drawing principally on a combination of STS and economic geography insights.

A theme that runs through all the literatures examined is the politics and social construction of policy success and failure. It is political science scholars who have examined this most closely, drawing our attention in particular to the temporal aspects of public policy development and the likelihood of policy successes or failures being promoted at certain moments. For example, the observation by Robertson (1991) and others that there are particular times in the policy process and political cycle (e.g. elections, policy committee deadlines) at which it becomes valuable to promote a policy as a failure or success, and/or when there is an urgent need to find policy solutions from elsewhere. There are, in other words, moments of change (so called ‘policy windows’; see Kingdon, 2003) where policy mobility is likely to be most active, and policy successes or failures promoted.

Within economic geography a key contribution to the consideration of policy failure is the distinction between tacit and codified knowledge. The tacit/codified categorization of knowledge-sharing and learning has been used extensively in economic geography to help explain processes of innovation and economic development, with a focus on corporations and other non-state actors. Empirical studies demonstrate that codified knowledge travels relatively easily, whereas tacit knowledge is less mobile. A key inference for the study of policy failure mobilities is the link between tacit knowledge and learning about failure. Economic geography scholars have, for example, observed how learning from failure tends to occur when actors are in close geographical proximity, through informal and spontaneous tacit forms of knowledge sharing (so-called ‘local buzz’; see Bathelt et al., 2004). In contrast, what is mobilized at broader spatial scales (nationally and internationally) tends to be positive successful case studies and examples, circulated using codified forms of knowledge. This observation alerts us to the possibility that, because policy mobility scholarship has focused primarily on international flows of policy, this is why there has been an empirical bias towards analysis of successful policies.

There are a number of ways in which these insights could be taken forward within policy mobilities research, and Table 1 summarizes their possible application. It outlines the different ways in which political science, economic geography and STS scholarship provide direction and insight for the study of policy failure mobilities. From this it is evident that the insights move beyond methodological remedies. In other words, the way forward involves not just better methodological balance and attention across policy successes and failures, but also further conceptual development within policy mobilities research. There are several directions this conceptual development could take. One starting point is to consider whether any of the insights from the three foundational areas of scholarship coalesce. The most promising research avenue in this regard is judged to be a focus on STS conceptualizations of fragility and materiality in order to empirically investigate the production and consumption of tacit and codified forms of policy knowledge. This would allow policy mobilities research to embrace an existing conceptual typology (tacit/codified knowledge) – used extensively in economic geography and beyond – in order to categorize and better understand the types of knowledge that are most important to the mobilization of policy successes and failures. Policy mobilities scholarship has to date been ambivalent about which type of knowledge is most important: there has been attention to tacit knowledge through the study of policy tourism (González, 2011) – via conferences, study tours and informal meeting places (airports, etc.) – regarding the circulation and informal bringing together of people and things. But there is also an opportunity to more systematically consider codified policy knowledge with policy mobilities research and the different forms it takes. Moreover, by combining this with attention to the materiality of mobile policies, and STS scholars’ insights about the fragility of (policy) assemblages, a research agenda investigating the materialities and mobilities of tacit and codified forms of mobile policy knowledge emerges. Such an approach would usefully move beyond a dichotomy of policy failure/success, allowing a richer exploration of how mobile policies are inherently a mix of success and failure.

Table 1.

A summary of key insights on policy failure mobilities from political science, economic geography and STS.

Area of scholarship	Key insights	Potential application in better understanding policy failure mobilities
Political science	Politics of framing a policy as a success or failure. Attention to the temporality of the policy process and key moments of decision-making. Exploration of different types of policy failure.	Investigation of the different types of work (practices, strategies, discourse) involved in each type of framing – success or failure – and who does this work; temporal enrichment of policy mobilities research, complementing its spatial focus, e.g. closer attention to ‘policy windows’ and how these mobilize policies, or are prompted (‘opened’) by mobile policy successes or failures; evaluating how the framing of a policy as either a success or failure might change over time, and the role of memory and materiality of infrastructure in this process; categorization of policy failure into process, programmatic and political failures.
Economic geography	Highlighting the importance of corporations and other non-state actors in processes of policy change. Categorization of codified and tacit knowledge, and their different mobilities/geographies.	Deeper investigation of the business/profit imperatives that create policy successes and foster their mobility, whilst tending to ‘close down’ and actively hamper policy failure mobilities; use of the tacit/codified typology to clarify different types of policy knowledge and their respective mobilities, e.g. to evaluate the extent to which knowledge about policy failures becomes codified (in reports, presentations, case study analysis), or remains tacit; examine the process of codification of policy knowledge to evaluate the relationship between highly mobile policy successes and their degree of codification – are these successes well known and mobile because they have been codified? Or, do research findings showing the importance of study tours and site visits indicate that tacit knowledge sharing is equally if not more important in encouraging policy mobilities than codified knowledge?; to better understand how policy tourism (site visits and travelling to places of policy success) encourages tacit knowledge sharing, including about things that did not work well.
STS	Inherent fragility of policy successes, and their tendency to unravel. The role of materiality in creating and eroding legitimacy and support for policies. Policy ‘immutable mobiles’ that are able to travel and stay intact.	Studies of policy failure mobilities could usefully be about small, partial (temporarily and spatially discrete) failures (‘unravellings’) of policy successes, not just about well-publicized failures, e.g. nuanced cases where global policy successes have elements of failure, or policies are viewed as failures in some localities and arenas, but as successes in others; closer investigation of the materiality of policy failure mobilities – what things (infrastructures, technologies, websites, reports, site visits, media stories) facilitate or hamper the mobilisation of policy failures? exploration of how the idea of an immutable mobile encapsulates how policy decision-makers tend to conceive of the borrowing of policies from elsewhere.

Further, there is a second, related research avenue on policy failure mobilities to be expanded on regarding the rich spatial description and geographical classification of failure. This draws inspiration from political science scholarship that views policy failure through a temporal lens – as process, programmatic and political failure, across the stages of the policy process – but takes geography rather than time as its starting point. Such a body of research would be focused on what moves and what does not in cases of policy failure: in some empirical cases of failure it is elements of policy success that have been immobilized (see Lovell, 2017), whereas in others it is failure (Webber, 2015). A geographical classification of policy failure would also allow for a fuller engagement with the hopes, fears and imaginations associated with policies – the topological spatialities of policy mobilities adeptly described by Robinson (2015). The application of a geographical lens to the classification of policy failure is judged to be an area of research that is of value, because it brings a spatial understanding of policy failure to the fore, and enables a richer appreciation of the complex interweaving of policy failure and success.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received funding from the Australian Research Council under their Future Fellowships Programme (Grant No. FT140100646; 2015-19) to support the publication of this article.

Note

References

Adey

Bissell

Hannam

Merriman

Sheller

(eds) (2014) The Routledge Handbook of Mobilities. London: Routledge.

Allen

Cochrane

(2007) Beyond the territorial fix: Regional assemblages, politics and power. Regional Studies 41: 1161–1175.

Baker

Smith

Roy

McCann

MGuirk

Mountz

Peck

Theodore

(2016) Reading Jamie Peck and Nik Theodore’s fast policy: Experimental statecraft at the thresholds of neoliberalism. Political Geography 53: 89–99.

Bathelt

(2006) Geographies of production: Growth regimes in spatial perspective 3 – toward a relational view of economic action and policy. Progress in Human Geography 30(2): 223–236.

Bathelt

Malmberg

Maskell

(2004) Clusters and knowledge: Local buzz, global pipelines and the process of knowledge creation. Progress in Human Geography 28(1): 31–56.

Bennett

Howlett

(1992) The lessons of learning: Reconciling theories of policy learning and policy change. Policy Sciences 25: 275–294.

Bovens

t’Hart

(1996) Understanding Policy Fiascos. New Brunswick, NJ: Transaction Books.

Bulkeley

(2006) Urban sustainability: Learning from best practice? Environment and Planning A 38: 1029–1044.

Bunnell

Coe

(2001) Spaces and scales of innovation. Progress in Human Geography 25(4): 569–589.

10.

Callon

(1986a) The sociology of an actor-network: The case of the electric vehicle. In: Callon

(ed.) Mapping the Dynamics of Science and Technology. London: Macmillan, 19–34.

11.

Callon

(1986) Some elements in a sociology of translation: Domestication of the scallops and fishermen of St. Brieuc Bay. Sociological Review Monograph 32: 196–233.

12.

Callon

(2007) What does it mean to say that economics is performative? In: MacKenzie

Muniesa

Siu

(eds) Do Economists Make Markets? On the Performativity of Economics. Princeton: Princeton University Press, 311–357.

13.

Clarke

(2012) Urban policy mobility, anti-politics, and histories of the transnational municipal movement. Progress in Human Geography 36(1): 25–43.

14.

Cochrane

Ward

(2012) Guest editorial: Researching the geographies of policy mobility: Confronting the methodological challenges. Environment and Planning A 44: 5–12.

15.

Cresswell

(2010) Mobilities I: Catching up. Progress in Human Geography 35: 550–558.

16.

Cresswell

(2012) Mobilities II: Still. Progress in Human Geography 36: 645–653.

17.

Deleuze

Guattari

(2004) A Thousand Plateaus, trans. Massumi

. London: Continuum.

18.

Didier

Peyroux

Morange

(2012) The spreading of the city improvement district model in Johannesburg and Cape Town: Urban regeneration and the neoliberal agenda in South Africa. International Journal of Urban and Regional Research 36: 915–935.

19.

Dolowitz

Marsh

(1996) Who learns what from whom: A review of the policy transfer literature. Political Studies 44: 343–357.

20.

Dolowitz

Marsh

(2000) Learning from abroad: The role of policy transfer in contemporary policy making. Governance 13: 5–24.

21.

Duncan

(2009) Policy transfer: Theory, rhetoric and reality. Policy & Politics 37: 453–458.

22.

Evans

Karvonen

(2014) ‘Give me a laboratory and I will lower your carbon footprint!’ Urban laboratories and the governance of low-carbon futures. International Journal of Urban and Regional Research 38: 413–430.

23.

Fløysand

Jakobsen

S-E

(2011) The complexity of innovation: A relational turn. Progress in Human Geography 35: 328–344.

24.

Geddie

(2015) Policy mobilities in the race for talent: Competitive state strategies in international student mobility. Transactions of the Institute of British Geographers 40: 235–248.

25.

Gertler

(1995) ‘Being there’: Proximity, organization, and culture in the development and adoption of advanced manufacturing technologies. Economic Geography 71: 1–26.

26.

Gertler

(2003) Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there). Journal of Economic Geography 3: 75–99.

27.

Gieryn

(2006) City as truth-spot: Laboratories and field-sites in urban studies. Social Studies of Science 36: 5–38.

28.

González

(2011) Bilbao and Barcelona ‘in motion’: How urban regeneration ‘models’ travel and mutate in the global flows of policy tourism. Urban Studies 48: 1397–1418.

29.

Graham

Marvin

(2001) Splintering Urbanism: Networked Infrastructures, Technological Mobilities and the Urban Condition. London: Routledge.

30.

Guy

Marvin

Moss

(2001) Infrastructure in Transition: Urban Networks, Buildings, Plans. London: Earthscan.

31.

Hardie

Mackenzie

(2007) Assembling an economic actor: The agencement of a hedge fund. The Sociological Review 55: 57–80.

32.

Higgins

Kitto

(2004) Mapping the dynamics of new forms of technological governance in agriculture: Methodological considerations. Environment and Planning A 36: 1397–1410.

33.

Howells

(2002) Tacit knowledge, innovation and economic geography. Urban Studies 39: 871–884.

34.

Howlett

(2012) The lessons of failure: Learning and blame avoidance in public policy-making. International Political Science Review 33: 539–555.

35.

Hudson

Bo-Yung

(2014) Policy transfer using the ‘gold standard’: Exploring policy tourism in practice. Policy & Politics 42: 495–511.

36.

Illical

Harrison

(2007) Protecting endangered species in the US and Canada: The role of negative lesson drawing. Canadian Journal of Political Science 40: 367–394.

37.

Jacobs

(2012) Urban geographies I: Still thinking cities relationally. Progress in Human Geography 36: 412–422.

38.

Jones

(2013) Geographies of production I: Relationality revisited and the ‘practice shift’ in economic geography. Progress in Human Geography 38(4): 605–615.

39.

Karvonen

Van Heur

(2014) Urban laboratories: Experiments in reworking cities. International Journal of Urban and Regional Research 38: 379–392.

40.

Kingdon

(2003) Agendas, Alternatives and Public Policies. New York: Harper Collins College Publishers.

41.

Larner

Laurie

(2010) Travelling technocrats, embodied knowledges: Globalising privatisation in telecoms and water. Geoforum 41: 218–226.

42.

Latour

(1987a) Opening one eye while closing the other…a note on some religious paintings. The Sociological Review 35: 15–38.

43.

Latour

(1987b) Science in Action. Cambridge, MA: Harvard University Press.

44.

Latour

Woolgar

(2013) Laboratory Life: The Construction of Scientific Facts. Princeton: Princeton University Press.

45.

Law

(1986) On the methods of long-distance control: Vessels, navigation and the Portuguese route to India. The Sociological Review. London: Routledge, 234–263.

46.

Law

(1987) Technology and heterogeneous engineering: The case of Portuguese expansion. In: Bijker

Hughes

Pinch

(eds) The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge, MA: MIT Press, 111–134.

47.

Law

Hassard

(1999) Actor Network Theory and After. Oxford: Blackwell & The Sociological Review.

48.

Law

Mol

(2001) Situating technoscience: An inquiry into spatialities. Environment and Planning D 19: 609–622.

49.

Leminen

Westerlund

Nyström

A-G

(2012) Living labs as open-innovation networks. Technology Innovation Management Review 2.

50.

Lovell

(2017) Are policy failures mobile? An investigation of the Advanced Metering Infrastructure Program in the State of Victoria, Australia. Environment and Planning A 49: 314–331.

51.

Lovell

Smith

(2010) Agencement in housing markets: The case of the UK construction industry. Geoforum 41: 457–468.

52.

MacKinnon

Cumbers

Chapman

(2002) Learning, innovation and regional development: A critical appraisal of recent debates. Progress in Human Geography 26: 293–311.

53.

Marsh

McConnell

(2010) Towards a framework for establishing policy success. Public Administration 88: 564–583.

54.

May

(1992) Policy learning and failure. Journal of Public Policy 12: 331–354.

55.

McCann

(2011) Urban policy mobilities and global circuits of knowledge: Toward a research agenda. Annals of the Association of American Geographers 101: 107–130.

56.

McCann

(2008) Expertise, truth, and urban policy mobilities: Global circuits of knowledge in the development of Vancouver, Canada’s ‘four pillar’ drug strategy. Environment and Planning A 40: 885–904.

57.

McCann

Ward

(2012) Policy assemblages, mobilities and mutations: Toward a multidisciplinary conversation. Political Studies Review 10: 325–332.

58.

McCann

Ward

(2015) Thinking through dualisms in urban policy mobilities. International Journal of Urban and Regional Research 39: 828–830.

59.

McFarlane

(2011) The city as a machine for learning. Transactions of the Institute of British Geographers 36: 360–376.

60.

Morgan

(2004) The exaggerated death of geography: Learning, proximity and territorial innovation systems. Journal of Economic Geography 4: 3–21.

61.

Mossberger

Wolman

(2003) Policy transfer as a form of prospective policy evaluation: Challenges and recommendations. Public Administration Review 63: 428–440.

62.

Müller

(2015) (Im-)mobile policies: Why sustainability went wrong in the 2014 Olympics in Sochi. European Urban and Regional Studies 22: 191–209.

63.

Peck

(2011) Geographies of policy: From transfer-diffusion to mobility-mutation. Progress in Human Geography 35: 773–797.

64.

Peck

Theodore

(2015) Fast Policy: Experimental Statecraft at the Thresholds of Neoliberalism. Minneapolis, US: The University of Minnesota Press.

65.

Polanyi

(1966) The Tacit Dimension. London: Routledge Kegan Paul.

66.

Pow

(2014) License to travel: Policy assemblage and the ‘Singapore model’. City 18: 287–306.

67.

Prince

(2012) Policy transfer, consultants and the geographies of governance. Progress in Human Geography 36: 188–203.

68.

Prince

(2014) Consultants and the global assemblage of culture and creativity. Transactions of the Institute of British Geographers 39: 90–101.

69.

Robertson

(1991) Political conflict and lesson-drawing. Journal of Public Policy 11: 55–78.

70.

Robinson

(2015) ‘Arriving at’ urban policies: The topological spaces of urban policy mobility. International Journal of Urban and Regional Research 39: 831–834.

71.

Rutten

(2016) Beyond proximities: The socio-spatial dynamics of knowledge creation. Progress in Human Geography 41: 159–177.

72.

Shapin

Schaffer

(1985) Leviathan and the Air Pump: Hobbes, Boyle and the Experimental Life. Princeton: Princeton University Press.

73.

Singleton

Michael

(1993) Actor-networks and ambivalence: General practitioners in the UK Cervical Screening Programme. Social Studies of Science 23: 227–264.

74.

Stone

(2004) Transfer agents and global networks in the ‘transnationalization’ of policy. Journal of European Public Policy 11: 545–566.

75.

Temenos

McCann

(2012) The local politics of policy mobility: Learning, persuasion, and the production of a municipal sustainability fix. Environment and Planning A 44: 1389–1406.

76.

Urry

(2007) Mobilities. Cambridge: Polity.

77.

VAGO (2015) Realising the Benefits of Smart Meters. Melbourne: Victorian Auditor-General’s Office (VAGO).

78.

Walker

(1969) The diffusion of innovations among the American states. American Political Science Review 63: 880–899.

79.

Ward

(2006) ‘Policies in motion’, urban management and state restructuring: The trans-local expansion of business improvement districts. International Journal of Urban and Regional Research 30: 54–75.

80.

Webber

(2015) Mobile adaptation and sticky experiments: Circulating best practices and lessons learned in climate change adaptation. Geographical Research 53: 26–38.

81.

Wood

(2015) The politics of policy circulation: Unpacking the relationship between South African and South American cities in the adoption of bus rapid transit. Antipode 47: 1062–1079.