— FROM TERRITORIAL COHESION TO REGIONAL SPATIAL JUSTICE: The Well-being of Future Generations Act in Wales

The European Union’s flagship Cohesion Policy faces evidence of dubious economic effectiveness and growing political and philosophical critiques of the very ideals of furthering European integration. This article examines ambitions for territorial cohesion as they have been operationalized through regional development in Wales. We argue that a potential alternative to the failed realization of territorial cohesion lies in the principles of spatial justice. While territorial cohesion has typically emphasized the redistribution of funds to ‘lagging’ regions, spatial justice, as we define it, is premised upon enabling regions to assert their own capacity to act and pursue positive visions of regional futures, consider the implications of space and scale for the achievement of justice, and define wellbeing, development and the ‘good life’ in ways that reflect regional priorities. We examine three ways in which recent political discourses and policy mechanisms in Wales resonate with these ideals. We focus on attempts to envision a progressive Welsh future, develop alternative spaces and scales of governance, and redefine and pluralize understandings of progress and well-being. The article concludes by reflecting on the practical and conceptual implications of rescaling spatial justice to regional contexts. Introduction The European Union’s (EU) Cohesion Policy is under pressure. An EU flagship policy since the late 1990s, Cohesion promised to ‘give to the people of Europe [a] unique blend of economic well-being, social cohesiveness and high overall quality of life’ (European Commission,  1994:1). Yet now the policy faces evidence of dubious economic effectiveness allied with growing political and philosophical critiques of the very ideals of furthering European integration. Ambitions for territorial cohesion appear particularly hollow given increasing economic disparities between European regions. Of the 26 countries with available data, 16 registered increases in the regional dispersion of GDP at NUTS3 level between 2007 and 2011, including France, Greece, Ireland, Italy, Poland, Spain and the United Kingdom (Eurostat,  2012). Economic disparities fuel more radical political narratives, in turn eroding the values behind Cohesion (González et al., 2015). Hence, against a background of rising inequality and austerity politics, populist parties loudly question the European project’s very viability. Left behind despite Cohesion’s promises, the ‘places that don’t matter’, as RodríguezPose (2018) provocatively suggests, may be taking their revenge. In this article, we examine––conceptually and empirically––whether the concept of spatial justice may offer a potential alternative to the failed promises of territorial cohesion. In recent years, several scholars have begun to advocate spatial justice as a means of understanding and alleviating socio-economic and political inequalities, primarily at the urban scale (e.g. Fainstein, 2010; Soja, 2010). This has, in turn, sparked interest in how spatial justice could be applied in regional or territorial contexts JONES, GOODWIN-HAWKINS AND WOODS 2 (e.g. Kearns et al., 2014; Lang and Görmar, 2019). While territorial cohesion has typically emphasized redistributing funds to ‘problem’ or ‘lagging’ regions, a regionalized spatial justice starts from the premise that regions should be able to: (1) assert their own capacity actively to define and pursue policy goals based on ideas of justice (cf. Malloy, 2010); (2) consider the implications of space and scale for the promotion of justice; (3) be able to define well-being, development and the ‘good life’ in ways that reflect regional priorities. Despite growing interest in the potential of spatial justice, there remains a need to examine both the conceptual and practical implications of regionalizing spatial justice, particularly in an EU context. In conceptual terms, there is a need to examine fully the implications of rescaling spatial justice from the urban to the regional scale. If, as is commonly accepted, space is more than merely a container for social processes of different kinds (Soja, 1989), then there is a need to reflect on the impact that defining spatial justice as a regional, a opposed to an urban, goal has on its meaning and operation. And there are also practical or policy-related implications to such a process of rescaling. To what extent does spatial justice allow regions to escape from dependency discourses and define progressive futures for themselves? To what extent does spatial justice enable regions to self-define progress, development and the ‘good life’? Further, how might the interface between existing approaches and emerging interventions reflect path dependency, and broader continuities and changes? Our aim in this article is to answer these significant conceptual and policy-related questions. We ground these concerns through an empirical focus on the case study of the region of Wales; one that has been subject to EU Cohesion Policy but is now trialling new policy goals embedded, we contend, in notions of spatial justice. Years of policy and academic discourse has problematized this small constituent country of the UK, marked by processes of post-industrial change and rural poverty, as a ‘lagging’ region. Years of EU funding has, in turn, promised to help Wales ‘catch up’ with more prosperous regions. In the 2014-20 EU programming period alone, Wales received 20% of the UK’s total structural funds allocation (Woolford and Hunt, 2016) for scarcely 5% of the population. Yet after almost two decades of inflated funding, the GDP map has barely changed for Wales and a Welsh majority voted to leave the EU in the Brexit referendum, with funding levels playing no predictive role in the vote (Becker et al., 2017).1 Furthermore, Wales is a region of the UK that is beginning to experiment with new modes of policy delivery; ones that are redolent, we suggest, of an approach grounded in notions of spatial justice. It is, therefore, a highly appropriate case study for examining the interplay between modalities of government informed by principles of territorial cohesion and those based on understandings of spatial justice. The article is structured as follows. In the next section, we review the literature on territorial cohesion and spatial justice. We particularly seek to highlight the distinctiveness of a spatial justice approach to addressing spatial inequalities at the regional scale. We then provide additional background on Wales as a case study, and detail the extensive empirical research underpinning our arguments. In the subsequent section, we discuss the recent transition in policy interventions in Wales from approaches reflecting territorial cohesion ideals to approaches that are, we believe, more illustrative of a desire to promote spatial justice. In particular, we identify recent political discourses and policy mechanisms in Wales that attempt to: highlight the Welsh region’s capacity to define a more just Welsh future; develop alternative regional and scalar frameworks that have the potential to promote spatial justice; redefine and pluralize understandings of progress and well-being. We conclude with reflections on the broader implications of shifting from more urban to more regional forms of spatial justice. 1 Public disillusionment with the EU occurred for numerous reasons. Moreover, the perceived ‘failure’ of EU regional investment may have been due to other factors, not least the macroeconomic policies of the UK, which have implicitly favoured certain regions above others. FROM TERRITORIAL COHESION TO REGIONAL SPATIAL JUSTICE 3 Reflecting on territorial cohesion and spatial justice — Making sense of territorial cohesion Territorial inequalities and uneven development have for some time been acknowledged as major concerns for Europe. Over the past two decades, the European Commission (2004: 27) has responded by promoting Cohesion Policy, understood as ‘a more balanced development [achieved] by reducing existing disparities, preventing territorial imbalances and making both sectoral policies which have a spatial impact and regional policy more coherent’. Thus, territorial cohesion can be read as a policy attempt to address spatialized issues that risk undermining the EU’s political and economic strength. In policy practice, such spatial issues are primarily understood as measurable differences in GDP between NUTS2 regions. Thereby defined by economic indicators, territorial cohesion is operationalized through financial mechanisms, including the European Social Fund and European Regional Development Fund. Variance in regional GDP is not necessarily considered problematic per se. Rather, there is a belief that the EU cannot afford the economic burden of ‘lagging’ regions, which threaten to diminish the overall competitiveness of the EU as a whole (Jones et al., 2019). That these regions are understood to contain underutilized or unrealized territorial potential is reflected in the labelling of cohesion funding as regional ‘investment’, rather than redistribution or charity. Beyond practical concerns, the concept of territorial cohesion has itself been the subject of considerable conceptual critique. First, scholars have described territorial cohesion as an ambiguous concept (e.g. Mirwaldt et al., 2008; Servillo, 2010; Atkinson and Zimmerman, 2016). Sometimes, it is unclear whether territorial cohesion is intended as the policy outcome or the policy itself, and the European Commission has remained reluctant to fully define what it actually hopes to realize through promoting cohesion. While this conceptual ‘fuzziness’ has certainly helped territorial cohesion become accepted across EU member states (Davoudi, 2005; Faludi, 2007), closer inspection suggests that states and regions may be simply manipulating the concept to suit their own purposes (e.g. Boland, 2005; Faludi, 2015; Van Well, 2012). A second critique relates to the tendency for territorial cohesion to render the EU into a singular geopolitical object to be measured, mapped, analysed and acted upon (Luukkonen and Moisio, 2016). Viewed as a political technology, territorial cohesion poses a challenge to nation-states’ identities, potentially rescaling territory––as a form of political calculation––so that it operates at a European rather than national scale (Elden, 2010). Third, there is potential for a discourse of territorial cohesion––especially when allied with talk of regional inequalities––to reinforce a perception that certain ‘underperforming’ regions are somehow problematic or lacking. But the problem with this view of some regions being stragglers in a race is that the race has already been run and the finish line drawn by stronger, faster sprinters (Hadjimichalis, 2019). There is a further danger of a status of inadequacy or even victimhood being ascribed to ‘lagging’ regions. Fourth, by focusing on particular economic and social measures of success or failure, territorial cohesion tends to discursively present development, well-being and the ‘good life’ in narrow ways, leaving little scope for alternative measures of progress to emerge. Again, limited visions of success or failure reinforce the notion that ‘lagging’ regions have to play a game, whose rules are defined elsewhere. Finally, territorial cohesion is predicated on problematic spatial imaginaries. While NUTS2 level GDP maps tell us an important story about regional disparities, their deployment in territorial cohesion policies effectively privileges some regions as the normative standard, while the laggards light up luridly, made problematic through an economic cartography always beyond their choosing (cf. González et al., 2015). Moreover, these maps highlight how particular geographies of governance have been established in different parts of Europe, often with the explicit aim of accessing EU funds. Yet, there is a danger that the regional geographies created in this way serve to JONES, GOODWIN-HAWKINS AND WOODS 4 reproduce the inequalities they were designed to ameliorate. There is certainly evidence that the two Welsh regions devised to access EU funds have confounded attempts to address socio-economic inequalities, and we return to this paradox shortly. We consider that the issues outlined so far highlight the need to move beyond territorial cohesion towards alternative approaches to addressing regional inequalities in Europe (Davoudi, 2007). Some evidence suggests that the EU is itself exploring alternatives. For example, Hadjimichalis (2019) claims that while territorial cohesion was high on the political agenda before the crisis of 2008-2010, it has now become totally downplayed. Hadjimichalis may be overstating matters, but there is clearly an appetite in academic and policy circles to develop alternative conceptualizations to territorial cohesion. Spatial justice may provide one such option. — Approaching spatial justice Spatial justice has been subject to academic and policy discussion in two main periods. Beginning in the late 1960s, the first period was marked by an interrogation of spatial justice by authors such as: Davies (1968), who attempted to understand how local services were distributed with respect to the needs of designated service areas; Lefebvre (1970), who conceptualized the fundamental socio-political changes needed to secure a spatially just society, and; Harvey (1973: 306), who viewed the city as not only the culmination of the spatial (in)justices associated with capitalism but also the beginnings of their abolition. The second period of spatial justice literature began in the new millennium, with the publication of seminal books including Fainstein’s (2010) The Just City––in which she argued for the redistribution of resources and opportunities, the celebration of diversity, and democracy––and Soja’s (2010) Seeking Spatial Justice, where the struggle over metropolitan transport in Los Angeles was used as a case study for a general argument about spatial rights and the processes that create injustice at the urban scale. The academic literature on spatial justice possesses key features. First, it is a concept that has been overwhelmingly approached through reference to cities and the urban scale. Cities have been viewed as manifestations of the economic inequalities that characterize modern society, as well as key sites for imagining alternative and just socio-spatial forms. Yet there is no necessary link between spatial justice and the urban scale. Merrifield and Swyngedouw (1997: 3) maintain that justice should take spatial, temporal and scalar differences into account. Similarly, Soja (2010: 20) observes that ‘justice and injustice are infused into the multi-scalar geographies in which we live, from the intimacies of the household to the uneven development of the global economy’. While we should not, then, view spatial justice as exclusively articulated through the urban scale, we need also recognize that spatial in/justices do not necessarily play out in the same ways across different spaces and scales. Consequently, we need to examine the implications of rescaling spatial justice from an urban to more regional, national and global contexts (Kearns and Reid-Henry, 2009; Kearns et al., 2014). Second, there is a need to interrogate the relationship between space and justice. Work in Geography and beyond has shown that space should not be viewed merely as a container for socio-economic processes (e.g. Soja, 1989). Consequently, Dabinett (2011: 2391) has argued that we should avoid seeing space as merely a container for justice, or spatial justice as ‘shorthand for social justice in space’. Rather, space needs to be interrogated to determine the various, far-reaching ways that it influences in/justice. Landscapes, for example, can be read as material and symbolic representations of past injustices, and as spatial contexts that help to reproduce inequalities (Mitchell, 2003: 788). Similarly, there is a need to consider how spaces of governance may reinforce inequalities and injustices of different kinds or allow for new approaches to justice to be imagined (Fainstein, 2001). Third, researchers have highlighted the significance of capacity for understandings of spatial justice, particularly through the ability of stakeholders to FROM TERRITORIAL COHESION TO REGIONAL SPATIAL JUSTICE 5 shape more just social and economic forms. Various authors approach justice from the perspective of capacity. Lefebvre (1970) and other contemporaries claimed that justice involved the right to take part in urban transformation processes, implying an ‘active participation in the political life, management, and the administration of the city’ (Dikeç, 2001: 1790). More recently, Sen’s (1993; 2009) capabilities approach has influenced Israel and Frenkel’s (2017: 2) notion of a justice deriving from a ‘person’s capabilities and ... liberties’. Such a focus on capacity at the individual and urban scale can be rescaled into a conception of spatial justice as reflecting a particular region’s capabilities and liberties to shape its own future. As we shall see in our case study, the capacity to define just futures moves us well beyond the kind of regional victimhood that is reinforced by territorial cohesion. Fourth, spatial justice should be understood in plural ways. While some universalist goals may be constant, authors like Sen (2009) argue that we must avoid imagining a perfectly just situation against which to measure reality in all spatial and temporal circumstances. Israel and Frenkel (2017) similarly insist that there is no single definition of justice: it should be assessed on the basis of individual opportunities in a given context, alongside plural understandings of what is fair and what may create wellbeing (Sen, 1993). Storper (2011: 19), too, maintains that although ‘freedom and liberty; the ability to live our lives and be happy; and [the] development of our capabilities’ may well be common goals, in practice ‘different individuals, groups and territories ... fill in the detail on these goals in rather different ways’. Thus, approaching spatial justice plurally points to the need to go beyond universalizing measures like GDP when seeking to define justice and well-being. Such sentiments echo the growing significance placed within public policy on seeking to promote well-being and happiness, not just wealth (e.g. Stiglitz et al., 2009), and the efforts of critical geographers to promote vital geographies, encompassing ‘fairness, care ... human rights, and solidarity with environmental and social justice’ (Kearns and Reid-Henry, 2009: 554). Overall, the above discussion highlights our significant and original conceptual contribution in this article; namely our explication of the key differences between conventional approaches to territorial cohesion and the literature on spatial justice. We have also begun to foreground some of the consequences of operationalizing the latter concept at the regional scale. While territorial cohesion tends towards discourses of victimhood and helplessness, we suggest that scaling spatial justice to the region emphasizes the need to develop regions’ capacities to transform themselves. Whereas territorial cohesion foregrounds a rather limited conceptualization of development and progress, spatial justice stresses plurality; hence regions should be able to define their own, specific, development routes. Finally, territorial cohesion has arguably imposed spatial and scalar straightjackets on regions, making the furthering of more equal and just societies more, not less, challenging. Spatial justice may potentially empower regions to define alternative spaces of governance that are required to promote spatial justice in effective ways. However, the potential of spatial justice signalled by the literature is still largely supposition, with little empirical work yet examining the effectiveness of such an alternative approach. As we move now to our case study, we contribute an original case study exploring the value of spatial justice in addressing socio-economic inequalities at a regional scale. Case study and methods Wales, for a number of reasons, presents a highly appropriate case through which to examine the conceptual consequences of a transition at a regional scale from territorial cohesion measures to an approach informed by spatial justice. First, Wales, as a region of the UK, has long been beset by challenging socioeconomic circumstances and it has sought to respond to them through targeted and, increasingly, bespoke policy interventions. Challenges include the contemporary JONES, GOODWIN-HAWKINS AND WOODS 6 malaise of an economy historically dominated by agriculture and heavy industry, a lowskilled and ageing population, poor infrastructures, and an economic geography offering ‘the least opportunity to benefit from “economic” mass of any UK country or region’ (Prince, 2016: 42). Since the devolution of power in 1999 through the creation of a Welsh Government, the case has been made for addressing entrenched ‘Welsh problems’ using distinct policy instruments (House of Commons, 1997). As Heley (2013: 1326) writes, devolution ‘has come with the understanding that longstanding social and economic disparities’ in Wales ‘cannot be adequately addressed through generic, UK-wide strategy and resource allocation’. The Welsh Government’s flagship programme to alleviate poverty in disadvantaged areas has been the recently disestablished Communities First scheme (Hincks and Robson, 2010), but other place-based schemes have sought to address socio-economic challenges in more targeted ways (e.g. Welsh Government, 2018). The key point here is that Wales has had, especially since 1999, an administrative capacity and political will to address socio-economic challenges, and Wales-specific policy mechanisms have been developed to address them (cf. Jones et al., 2005). Second, ‘Welsh problems’ have enabled certain parts of Wales to qualify for EU funding. As Royles (2006: 138) indicates, by ‘unique coincidence’ the NUTS2 region West Wales and the Valleys was first designated as what was then called an ‘Objective 1’ area in 2000, just a year after executive devolution had occurred in Wales. As ‘[p]laques bearing the “Funded by the European Union” logo became nigh on ubiquitous’ (Wyn Jones and Rumbul, 2013: 559), the promise of the increased EU funding flows associated with Objective 1 status brought ‘cautious optimism that the problems that had marked out Wales as one of the poorest regions in the EU could now be addressed’ (Fudge, 2006: 55). A substantial amount of EU funding has indeed been allocated to Wales since 2000, with €2.4 billion allocated through the ERDF and ESF during 2014-20 alone (Woolford and Hunt, 2016). West Wales and the Valleys (see Figure 1) is one of only two NUTS2 regions in the UK calculated to be net recipients of EU funds, equating to around £79 or €90 per capita (Ifan et al., 2016). As we show below, regional development in Wales has been characterized by both hope for the potential promised by EU Territorial Cohesion and considerable disillusionment in terms of its actual impact. Third, the Welsh Government has committed to promote sustainable development, well-being and justice (Bishop and Flynn, 2004). Section 121 of the Government of Wales Act 1998 required the Welsh Government ‘to make a scheme setting out how it proposes, in the exercise of its functions, to promote sustainable development’. Most recently, this commitment has been manifest in the Well-being of Future Generations Act 2015. The Act proceeded from a national conversation on ‘The Wales We Want’ by 2050 (Jones and Ross, 2016), and reflects an attempt by the Welsh Government to collaborate with stakeholders across Wales to develop a distinctive approach to sustainable development. This legislation is particularly significant given the emphasis it places on, first, redefining sustainable development as a concern with well-being and justice, and, second, defining the ‘good life’ in Wales in varied and inclusive ways, which extend beyond economic measures of success (see also Welsh Government, 2011: iv). While some of this policy emphasis may reflect the Welsh Government’s need ‘to demonstrate that devolution [is] “making a difference”’ (Royles, 2006: 147), we contend that it also represents a significant attempt to define the future for Wales in progressive and distinctive ways, and to envisage more sociospatially just outcomes. Finally, the imminent departure (at the time of writing) of the UK from the EU makes the discussion about the transition from EU funding in Wales even more pertinent. Since the Brexit referendum, attempts have been made to define replacement funding mechanisms for Wales. The Welsh Government’s (2017a: 7) Regional Investment in Wales After Brexit insists that ‘[r]egional disparities remain a challenge and we can only address them by setting a national strategy underpinned by regionally designed and FROM TERRITORIAL COHESION TO REGIONAL SPATIAL JUSTICE 7 managed action plans’. While there are echoes of territorial cohesion in sentences such as ‘Investments must be targeted geographically’ (ibid., our emphasis), the white paper begins to shift the discourse from the imminent loss of funding towards the potential for new ways of recognizing the ‘long-term challenges’ facing Wales, along with the ‘need for a more granular understanding of the specific needs of different places’ (ibid.: 13). Although the contours of these new strategies are still opaque, the reality of Brexit clearly points to an opportunity for Wales to redefine ways of addressing regional inequalities; ones that extend beyond European-level concerns with territorial cohesion. The empirical material we discuss below focuses on these issues. Our data forms part of a broader Horizon 2020 project, which brings together 16 partners from 0 N


Introduction
The European Union's (EU) Cohesion Policy is under pressure. An EU flagship policy since the late 1990s, Cohesion promised to 'give to the people of Europe [a] unique blend of economic well-being, social cohesiveness and high overall quality of life' (European Commission, 1994:1). Yet now the policy faces evidence of dubious economic effectiveness allied with growing political and philosophical critiques of the very ideals of furthering European integration. Ambitions for territorial cohesion appear particularly hollow given increasing economic disparities between European regions. Of the 26 countries with available data, 16 registered increases in the regional dispersion of GDP at NUTS3 level between 2007 and 2011, including France, Greece, Ireland, Italy, Poland, Spain and the United Kingdom (Eurostat, 2012). Economic disparities fuel more radical political narratives, in turn eroding the values behind Cohesion (González et al., 2015). Hence, against a background of rising inequality and austerity politics, populist parties loudly question the European project's very viability. Left behind despite Cohesion's promises, the 'places that don't matter', as Rodríguez-Pose (2018) provocatively suggests, may be taking their revenge.
In this article, we examine--conceptually and empirically--whether the concept of spatial justice may offer a potential alternative to the failed promises of territorial cohesion. In recent years, several scholars have begun to advocate spatial justice as a means of understanding and alleviating socio-economic and political inequalities, primarily at the urban scale (e.g. Fainstein, 2010;Soja, 2010). This has, in turn, sparked interest in how spatial justice could be applied in regional or territorial contexts (e.g. Kearns et al., 2014;Lang and Görmar, 2019). While territorial cohesion has typically emphasized redistributing funds to 'problem' or 'lagging' regions, a regionalized spatial justice starts from the premise that regions should be able to: (1) assert their own capacity actively to define and pursue policy goals based on ideas of justice (cf. Malloy, 2010); (2) consider the implications of space and scale for the promotion of justice; (3) be able to define well-being, development and the 'good life' in ways that reflect regional priorities. Despite growing interest in the potential of spatial justice, there remains a need to examine both the conceptual and practical implications of regionalizing spatial justice, particularly in an EU context. In conceptual terms, there is a need to examine fully the implications of rescaling spatial justice from the urban to the regional scale. If, as is commonly accepted, space is more than merely a container for social processes of different kinds (Soja, 1989), then there is a need to reflect on the impact that defining spatial justice as a regional, a opposed to an urban, goal has on its meaning and operation. And there are also practical or policy-related implications to such a process of rescaling.
To what extent does spatial justice allow regions to escape from dependency discourses and define progressive futures for themselves? To what extent does spatial justice enable regions to self-define progress, development and the 'good life'? Further, how might the interface between existing approaches and emerging interventions reflect path dependency, and broader continuities and changes? Our aim in this article is to answer these significant conceptual and policy-related questions.
We ground these concerns through an empirical focus on the case study of the region of Wales; one that has been subject to EU Cohesion Policy but is now trialling new policy goals embedded, we contend, in notions of spatial justice. Years of policy and academic discourse has problematized this small constituent country of the UK, marked by processes of post-industrial change and rural poverty, as a 'lagging' region. Years of EU funding has, in turn, promised to help Wales 'catch up' with more prosperous regions. In the 2014-20 EU programming period alone, Wales received 20% of the UK's total structural funds allocation (Woolford and Hunt, 2016) for scarcely 5% of the population. Yet after almost two decades of inflated funding, the GDP map has barely changed for Wales and a Welsh majority voted to leave the EU in the Brexit referendum, with funding levels playing no predictive role in the vote (Becker et al., 2017). 1 Furthermore, Wales is a region of the UK that is beginning to experiment with new modes of policy delivery; ones that are redolent, we suggest, of an approach grounded in notions of spatial justice. It is, therefore, a highly appropriate case study for examining the interplay between modalities of government informed by principles of territorial cohesion and those based on understandings of spatial justice.
The article is structured as follows. In the next section, we review the literature on territorial cohesion and spatial justice. We particularly seek to highlight the distinctiveness of a spatial justice approach to addressing spatial inequalities at the regional scale. We then provide additional background on Wales as a case study, and detail the extensive empirical research underpinning our arguments. In the subsequent section, we discuss the recent transition in policy interventions in Wales from approaches reflecting territorial cohesion ideals to approaches that are, we believe, more illustrative of a desire to promote spatial justice. In particular, we identify recent political discourses and policy mechanisms in Wales that attempt to: highlight the Welsh region's capacity to define a more just Welsh future; develop alternative regional and scalar frameworks that have the potential to promote spatial justice; redefine and pluralize understandings of progress and well-being. We conclude with reflections on the broader implications of shifting from more urban to more regional forms of spatial justice.

1
Public disillusionment with the EU occurred for numerous reasons. Moreover, the perceived 'failure' of EU regional investment may have been due to other factors, not least the macroeconomic policies of the UK, which have implicitly favoured certain regions above others.
Reflecting on territorial cohesion and spatial justice -

Making sense of territorial cohesion
Territorial inequalities and uneven development have for some time been acknowledged as major concerns for Europe. Over the past two decades, the European Commission (2004: 27) has responded by promoting Cohesion Policy, understood as 'a more balanced development [achieved] by reducing existing disparities, preventing territorial imbalances and making both sectoral policies which have a spatial impact and regional policy more coherent'. Thus, territorial cohesion can be read as a policy attempt to address spatialized issues that risk undermining the EU's political and economic strength. In policy practice, such spatial issues are primarily understood as measurable differences in GDP between NUTS2 regions. Thereby defined by economic indicators, territorial cohesion is operationalized through financial mechanisms, including the European Social Fund and European Regional Development Fund. Variance in regional GDP is not necessarily considered problematic per se. Rather, there is a belief that the EU cannot afford the economic burden of 'lagging' regions, which threaten to diminish the overall competitiveness of the EU as a whole . That these regions are understood to contain underutilized or unrealized territorial potential is reflected in the labelling of cohesion funding as regional 'investment', rather than redistribution or charity.
Beyond practical concerns, the concept of territorial cohesion has itself been the subject of considerable conceptual critique. First, scholars have described territorial cohesion as an ambiguous concept (e.g. Mirwaldt et al., 2008;Servillo, 2010;Atkinson and Zimmerman, 2016). Sometimes, it is unclear whether territorial cohesion is intended as the policy outcome or the policy itself, and the European Commission has remained reluctant to fully define what it actually hopes to realize through promoting cohesion. While this conceptual 'fuzziness' has certainly helped territorial cohesion become accepted across EU member states (Davoudi, 2005;Faludi, 2007), closer inspection suggests that states and regions may be simply manipulating the concept to suit their own purposes (e.g. Boland, 2005;Faludi, 2015;Van Well, 2012). A second critique relates to the tendency for territorial cohesion to render the EU into a singular geopolitical object to be measured, mapped, analysed and acted upon (Luukkonen and Moisio, 2016). Viewed as a political technology, territorial cohesion poses a challenge to nation-states' identities, potentially rescaling territory--as a form of political calculation--so that it operates at a European rather than national scale (Elden, 2010).
Third, there is potential for a discourse of territorial cohesion--especially when allied with talk of regional inequalities--to reinforce a perception that certain 'underperforming' regions are somehow problematic or lacking. But the problem with this view of some regions being stragglers in a race is that the race has already been run and the finish line drawn by stronger, faster sprinters (Hadjimichalis, 2019). There is a further danger of a status of inadequacy or even victimhood being ascribed to 'lagging' regions. Fourth, by focusing on particular economic and social measures of success or failure, territorial cohesion tends to discursively present development, well-being and the 'good life' in narrow ways, leaving little scope for alternative measures of progress to emerge. Again, limited visions of success or failure reinforce the notion that 'lagging' regions have to play a game, whose rules are defined elsewhere.
Finally, territorial cohesion is predicated on problematic spatial imaginaries. While NUTS2 level GDP maps tell us an important story about regional disparities, their deployment in territorial cohesion policies effectively privileges some regions as the normative standard, while the laggards light up luridly, made problematic through an economic cartography always beyond their choosing (cf. González et al., 2015). Moreover, these maps highlight how particular geographies of governance have been established in different parts of Europe, often with the explicit aim of accessing EU funds. Yet, there is a danger that the regional geographies created in this way serve to reproduce the inequalities they were designed to ameliorate. There is certainly evidence that the two Welsh regions devised to access EU funds have confounded attempts to address socio-economic inequalities, and we return to this paradox shortly.
We consider that the issues outlined so far highlight the need to move beyond territorial cohesion towards alternative approaches to addressing regional inequalities in Europe (Davoudi, 2007). Some evidence suggests that the EU is itself exploring alternatives. For example, Hadjimichalis (2019) claims that while territorial cohesion was high on the political agenda before the crisis of 2008-2010, it has now become totally downplayed. Hadjimichalis may be overstating matters, but there is clearly an appetite in academic and policy circles to develop alternative conceptualizations to territorial cohesion. Spatial justice may provide one such option.

-
Approaching spatial justice Spatial justice has been subject to academic and policy discussion in two main periods. Beginning in the late 1960s, the first period was marked by an interrogation of spatial justice by authors such as: Davies (1968), who attempted to understand how local services were distributed with respect to the needs of designated service areas; Lefebvre (1970), who conceptualized the fundamental socio-political changes needed to secure a spatially just society, and; Harvey (1973: 306), who viewed the city as not only the culmination of the spatial (in)justices associated with capitalism but also the beginnings of their abolition. The second period of spatial justice literature began in the new millennium, with the publication of seminal books including Fainstein's (2010) The Just City--in which she argued for the redistribution of resources and opportunities, the celebration of diversity, and democracy--and Soja's (2010) Seeking Spatial Justice, where the struggle over metropolitan transport in Los Angeles was used as a case study for a general argument about spatial rights and the processes that create injustice at the urban scale.
The academic literature on spatial justice possesses key features. First, it is a concept that has been overwhelmingly approached through reference to cities and the urban scale. Cities have been viewed as manifestations of the economic inequalities that characterize modern society, as well as key sites for imagining alternative and just socio-spatial forms. Yet there is no necessary link between spatial justice and the urban scale. Merrifield and Swyngedouw (1997: 3) maintain that justice should take spatial, temporal and scalar differences into account. Similarly, Soja (2010: 20) observes that 'justice and injustice are infused into the multi-scalar geographies in which we live, from the intimacies of the household to the uneven development of the global economy'. While we should not, then, view spatial justice as exclusively articulated through the urban scale, we need also recognize that spatial in/justices do not necessarily play out in the same ways across different spaces and scales. Consequently, we need to examine the implications of rescaling spatial justice from an urban to more regional, national and global contexts (Kearns and Reid-Henry, 2009;Kearns et al., 2014).
Second, there is a need to interrogate the relationship between space and justice. Work in Geography and beyond has shown that space should not be viewed merely as a container for socio-economic processes (e.g. Soja, 1989). Consequently, Dabinett (2011Dabinett ( : 2391 has argued that we should avoid seeing space as merely a container for justice, or spatial justice as 'shorthand for social justice in space'. Rather, space needs to be interrogated to determine the various, far-reaching ways that it influences in/justice. Landscapes, for example, can be read as material and symbolic representations of past injustices, and as spatial contexts that help to reproduce inequalities (Mitchell, 2003: 788). Similarly, there is a need to consider how spaces of governance may reinforce inequalities and injustices of different kinds or allow for new approaches to justice to be imagined (Fainstein, 2001).
Third, researchers have highlighted the significance of capacity for understandings of spatial justice, particularly through the ability of stakeholders to shape more just social and economic forms. Various authors approach justice from the perspective of capacity. Lefebvre (1970) and other contemporaries claimed that justice involved the right to take part in urban transformation processes, implying an 'active participation in the political life, management, and the administration of the city' (Dikeç, 2001(Dikeç, : 1790. More recently, Sen's (1993;2009) capabilities approach has influenced Israel and Frenkel's (2017: 2) notion of a justice deriving from a 'person's capabilities and … liberties'. Such a focus on capacity at the individual and urban scale can be rescaled into a conception of spatial justice as reflecting a particular region's capabilities and liberties to shape its own future. As we shall see in our case study, the capacity to define just futures moves us well beyond the kind of regional victimhood that is reinforced by territorial cohesion.
Fourth, spatial justice should be understood in plural ways. While some universalist goals may be constant, authors like Sen (2009) argue that we must avoid imagining a perfectly just situation against which to measure reality in all spatial and temporal circumstances. Israel and Frenkel (2017) similarly insist that there is no single definition of justice: it should be assessed on the basis of individual opportunities in a given context, alongside plural understandings of what is fair and what may create wellbeing (Sen, 1993). Storper (2011: 19), too, maintains that although 'freedom and liberty; the ability to live our lives and be happy; and [the] development of our capabilities' may well be common goals, in practice 'different individuals, groups and territories … fill in the detail on these goals in rather different ways'. Thus, approaching spatial justice plurally points to the need to go beyond universalizing measures like GDP when seeking to define justice and well-being. Such sentiments echo the growing significance placed within public policy on seeking to promote well-being and happiness, not just wealth (e.g. Stiglitz et al., 2009), and the efforts of critical geographers to promote vital geographies, encompassing 'fairness, care … human rights, and solidarity with environmental and social justice' (Kearns and Reid-Henry, 2009: 554).
Overall, the above discussion highlights our significant and original conceptual contribution in this article; namely our explication of the key differences between conventional approaches to territorial cohesion and the literature on spatial justice. We have also begun to foreground some of the consequences of operationalizing the latter concept at the regional scale. While territorial cohesion tends towards discourses of victimhood and helplessness, we suggest that scaling spatial justice to the region emphasizes the need to develop regions' capacities to transform themselves. Whereas territorial cohesion foregrounds a rather limited conceptualization of development and progress, spatial justice stresses plurality; hence regions should be able to define their own, specific, development routes. Finally, territorial cohesion has arguably imposed spatial and scalar straightjackets on regions, making the furthering of more equal and just societies more, not less, challenging. Spatial justice may potentially empower regions to define alternative spaces of governance that are required to promote spatial justice in effective ways. However, the potential of spatial justice signalled by the literature is still largely supposition, with little empirical work yet examining the effectiveness of such an alternative approach. As we move now to our case study, we contribute an original case study exploring the value of spatial justice in addressing socio-economic inequalities at a regional scale.

Case study and methods
Wales, for a number of reasons, presents a highly appropriate case through which to examine the conceptual consequences of a transition at a regional scale from territorial cohesion measures to an approach informed by spatial justice.
First, Wales, as a region of the UK, has long been beset by challenging socioeconomic circumstances and it has sought to respond to them through targeted and, increasingly, bespoke policy interventions. Challenges include the contemporary malaise of an economy historically dominated by agriculture and heavy industry, a lowskilled and ageing population, poor infrastructures, and an economic geography offering 'the least opportunity to benefit from "economic" mass of any UK country or region' (Prince, 2016: 42). Since the devolution of power in 1999 through the creation of a Welsh Government, the case has been made for addressing entrenched 'Welsh problems' using distinct policy instruments (House of Commons, 1997). As Heley (2013Heley ( : 1326 writes, devolution 'has come with the understanding that longstanding social and economic disparities' in Wales 'cannot be adequately addressed through generic, UK-wide strategy and resource allocation'. The Welsh Government's flagship programme to alleviate poverty in disadvantaged areas has been the recently disestablished Communities First scheme (Hincks and Robson, 2010), but other place-based schemes have sought to address socio-economic challenges in more targeted ways (e.g. Welsh Government, 2018).
The key point here is that Wales has had, especially since 1999, an administrative capacity and political will to address socio-economic challenges, and Wales-specific policy mechanisms have been developed to address them (cf. Jones et al., 2005).
Second, 'Welsh problems' have enabled certain parts of Wales to qualify for EU funding. As Royles (2006: 138) indicates, by 'unique coincidence' the NUTS2 region West Wales and the Valleys was first designated as what was then called an 'Objective 1' area in 2000, just a year after executive devolution had occurred in Wales. As '[p]laques bearing the "Funded by the European Union" logo became nigh on ubiquitous' (Wyn Jones and Rumbul, 2013: 559), the promise of the increased EU funding flows associated with Objective 1 status brought 'cautious optimism that the problems that had marked out Wales as one of the poorest regions in the EU could now be addressed' (Fudge, 2006: 55). A substantial amount of EU funding has indeed been allocated to Wales since 2000, with €2.4 billion allocated through the ERDF and ESF during 2014-20 alone (Woolford and Hunt, 2016). West Wales and the Valleys (see Figure 1) is one of only two NUTS2 regions in the UK calculated to be net recipients of EU funds, equating to around £79 or €90 per capita (Ifan et al., 2016). As we show below, regional development in Wales has been characterized by both hope for the potential promised by EU Territorial Cohesion and considerable disillusionment in terms of its actual impact.
Third, the Welsh Government has committed to promote sustainable development, well-being and justice (Bishop and Flynn, 2004). Section 121 of the Government of Wales Act 1998 required the Welsh Government 'to make a scheme setting out how it proposes, in the exercise of its functions, to promote sustainable development'. Most recently, this commitment has been manifest in the Well-being of Future Generations Act 2015. The Act proceeded from a national conversation on 'The Wales We Want' by 2050 (Jones and Ross, 2016), and reflects an attempt by the Welsh Government to collaborate with stakeholders across Wales to develop a distinctive approach to sustainable development. This legislation is particularly significant given the emphasis it places on, first, redefining sustainable development as a concern with well-being and justice, and, second, defining the 'good life' in Wales in varied and inclusive ways, which extend beyond economic measures of success (see also Welsh Government, 2011: iv). While some of this policy emphasis may reflect the Welsh Government's need 'to demonstrate that devolution [is] "making a difference"' (Royles, 2006: 147), we contend that it also represents a significant attempt to define the future for Wales in progressive and distinctive ways, and to envisage more sociospatially just outcomes.
Finally, the imminent departure (at the time of writing) of the UK from the EU makes the discussion about the transition from EU funding in Wales even more pertinent. Since the Brexit referendum, attempts have been made to define replacement funding mechanisms for Wales. The Welsh Government's (2017a: 7) Regional Investment in Wales After Brexit insists that '[r]egional disparities remain a challenge and we can only address them by setting a national strategy underpinned by regionally designed and managed action plans'. While there are echoes of territorial cohesion in sentences such as 'Investments must be targeted geographically' (ibid., our emphasis), the white paper begins to shift the discourse from the imminent loss of funding towards the potential for new ways of recognizing the 'long-term challenges' facing Wales, along with the 'need for a more granular understanding of the specific needs of different places' (ibid.: 13). Although the contours of these new strategies are still opaque, the reality of Brexit clearly points to an opportunity for Wales to redefine ways of addressing regional inequalities; ones that extend beyond European-level concerns with territorial cohesion.
The empirical material we discuss below focuses on these issues. Our data forms part of a broader Horizon 2020 project, which brings together 16 partners from 13 European states. The project aims to develop new policy mechanisms to enable European, national and regional governance agencies to address territorial inequalities more effectively. Importantly, the project also seeks to envision a future for Europe in which the distribution of resources is consistent with principles of spatial justice. To this end, one phase of research has developed a series of regional case studies within Europe, of which Wales is one. To date, a large amount of data has been collected and analysed, including: 69 expert interviews conducted in 2017-18 in six EU member states (Finland, Germany, Greece, Italy, Ireland and the UK); and, documentary research on public speeches and regional policy and cohesion reports published between 2004 and 2017. In interviews and through documentary research, we were interested in how policymakers conceptualized territorial inequalities, both within their regions and across the EU. We examined how ideas of territorial cohesion and spatial justice (and closely associated terms such as 'well-being' and the 'good life') are being used.
We discuss one subset of IMAJINE's data here, drawing from qualitative data collected through semi-structured expert interviews with Welsh policy actors. We interviewed senior civil servants and advisors from the devolved Welsh Government, and senior figures from local government in Wales. We aimed for a selective sample of respondents, covering a range of policy sectors, including education, health and transport (Coyne, 1997). We also draw on documentary evidence, namely Welsh Government policies, strategies and statements issued over the past 10-15 years. We were especially interested in those documents that sought to describe Wales' commitment--either independently or in concert with the EU--to addressing regional socio-spatial inequalities. We transcribed and coded the interview transcripts with NVivo, using a mixture of codes that were data-generated and others reflecting key conceptual themes (Coffey and Atkinson, 1996). The coding framework was agreed by the project team to ensure methodological rigour and consistency (Adair and Pastori, 2011). We used rapid appraisal techniques to identify and initially analyse documentary evidence (Kuchartz, 2014), noting common patterns within and between documents. Our goal in the empirical discussion below has been to use quotes that reflect dominant themes in the interviews and in the documentary evidence, as well as noting areas where alternative or contradictory themes emerged.

From territorial cohesion to spatial justice in Wales
Three key themes from our data illustrate the tentative transition in Wales from an approach to regional inequalities based on the logics of territorial cohesion, towards an alternative that is, we maintain, increasingly resonant with the principles of spatial justice. We discuss how: (1) a narrative of victimhood is changing into one of regional capacity; (2) more meaningful spaces and scales of governance are replacing instrumentally-defined regions; and, (3) multiple measures of progress and well-being are gaining salience over a fixation on GDP.
-From dependency to capacities to act Many of our interview respondents bemoaned the role that territorial cohesion has played in sustaining a discourse of dependence and victimhood in Wales. For example, an interviewee who joined the then-new Welsh civil service around the time of devolution and the allocation of the first tranche of structural funds recalled a 'great fanfare that this [EU funding] … was going to open up our communities … would get them out of poverty' (senior civil servant, governance). Yet, the same interviewee reflected on the situation at the end of the 2000-07 programming period: [The Welsh] Government, I think, was in quite a difficult position whereby our GDP per capita in [West Wales and the Valleys] hadn't really changed and we were still … below the European threshold. So, on the one hand this was a bit of an indictment of all the money. Has the money been wasted? On the other, there was this [sense that] we still actually want to stay under the threshold because then we'll qualify for another round of funding (senior civil servant, governance).
The last sentence of the quote above particularly shows how dependence developed in relation to cohesion funding. An attitude that there was value in staying 'under the threshold' took hold, partly due to the funding associated with maintaining Objective 1 status, and also because of the implications of a drop in funding for the jobs of individuals working for the Welsh Government, local authorities, and related agencies.
Once instrumentalized, the discourse of dependence and victimhood took on a performative and self-fulfilling character. Our data show the discourse operating at two distinct scales. As the comments above suggest, the discourse existed at a regional scale. Interviewees at their bluntest told us that, if EU funding really 'worked', then West Wales and the Valleys would not have kept qualifying for it. Despite years of funding, a senior economic advisor concluded, 'we started with a position of economic and social inequality and we certainly haven't closed that gap'. Indeed, almost all our interviewees believed that EU funding had proved ultimately ineffective in Wales. Unsurprisingly, then, interviewees did not afford the 2014-20 structural funds allocation of €2.4 billion to Wales (Woolford and Hunt, 2016) the same 'fanfare' that had characterized the first years of funding. By this time, Wales as a whole had assumed 'failing region' status, and was no longer invoked as near imminent rescue. Rather, economic underperformance and socio-economic inequalities were discursively entrenched as issues 'that bedevil us' (senior advisor, economy).
A spatialized narrative of failure, dependency and victimhood also operated at smaller scales. For our interviewees, post-industrial and rural areas, particularly, served as spatial synonyms for inequality and deprivation, with both kinds of spaces viewed as distinctly 'left behind' by progress elsewhere. While EU funding programmes ostensibly promise to combat such uneven development within regions, in reality, economic activity in Wales continues to largely concentrate on 'two main economic arteries' (senior advisor, economy) along the southern and northern coastline, generating growth that other Welsh communities simply do not benefit from. An interviewee reflected: Cardiff [the Welsh capital] as a city has seen some of the highest levels of economic growth and population growth compared to … any city in the UK. But you go 10 miles, 20 miles north [and] it hasn't changed in 30 years (senior advisor, economy).
When interviewees described travelling short distances--'10 miles, 20 miles'--and witnessing a perceptible socio-economic shift, they were invariably evoking a postindustrial or rural area. A senior civil servant in health observed that, in Wales, 'looking at geographies … tells a story about the inequity and the unfairness that people can buy into'. Such comments illustrate how certain spaces have taken on a particular status within the geographical imagination of politicians and policymakers alike. Localized inequalities have, thus, been narrated through: relational and topological understandings of space and distance (cf. Amin, 2004); discourses that conflate social and spatial inequalities (Townsend, 1979).
By describing the dependence and victimhood fostered through territorial cohesion approaches as a 'narrative', we do not mean that the socio-economic challenges bedevilling Wales lack substance. Real and objective challenges exist within and across Wales, as our interviewees were keen to point out, further criticizing the inability of EU funds to systematically address tangible inequalities: 'how do you deal with some of those areas of deprivation and those real inequalities?' (senior civil servant, governance). Our point here is that such inequalities have been consequently reworked into pervasive discourses of failure and dependence, due partly to importing territorial cohesion framings into Wales. As one policymaker put it: we keep talking about poverty. We keep talking about low skills. We keep talking about poor housing. And we keep coming up with programmes to fix it, but we never do (senior civil servant, governance).
While the inability to fundamentally address these issues may be because they are 'wicked problems', we also suspect that part of the narrative hinges on policymakers, essentially, 'talking about' these issues as problems. Territorial cohesion invites the perversity of 'lagging' regions putting their failures on display. In Wales, the result has been a discourse of dependence, leading to little more than developmental stasis; EU funding has 'basically just sustained where we were at' (senior advisor, local government).
Yet against this backdrop, new kinds of narrative have emerged. For the Welsh Government (2017a: 15), the UK's departure from the EU provides opportunities to 'think differently and work differently in ways which link policies together across portfolios and organisations'. Accordingly, the Welsh Government is attempting to think more pro-actively about long-term regional futures, instead of bemoaning a past and present characterized by injustice. We identify growing attention towards, first, envisaging more just and hopeful futures for Wales; and, second, developing regional capacities to shape those futures.
The Well-being of Future Generations Act 2015 is now the main vehicle for articulating a new vision for Wales. The Act seeks to mainstream sustainable development across all public bodies in Wales. Through an extensive consultation exercise, the Act also attempted to attune sustainable development ideals to the perceived priorities of the Welsh population (Jones and Ross, 2016), informing seven well-being goals (Welsh Government, 2015a: 3). The consultation envisioned a future Wales--'The Wales We Want'--which would be more sustainable and more just. In a startlingly ambitious statement, the Welsh Government's own submission to the consultation process insisted that: In 2050, Wales will be the best place to live, learn, work and do business … Doing things differently is about looking forward so the choices we take secure a safe and prosperous future for us, for our children and for our grandchildren (Welsh Government, 2014: 1).
This ambitious language differs considerably from the dependence discourse we discussed above. Caution is of course needed in attempting to compare competing discourses, since we refer to different statements made for different audiences. Yet, we do consider that the language associated with the Well-being of Future Generations Act represents a shift in the Welsh Government's approach to regional inequalities. Rather than a geography of problems and failure, the new well-being language attends to more progressive future goals. Rather than emphasizing past failure in expectation of future funds, present problems become inspiration for future action; as an interviewee involved in the Act's development told us, 'Wales has some of the highest rates of obesity in the world, some of the highest rates of teenage pregnancy … we need to change that as we're going forward' (senior advisor, Sustain Wales).
This shift in rhetoric is significant in its own right but there is also growing evidence that the emergence of this new agenda is having an impact on the ground in different public bodies in Wales. In one local authority in south Wales, respondents stressed that the Well-being Act had led to a change of emphasis in policy discussions, with, a more holistic take on well-being taking root, and one that emphasized an apiration to create better standards of living for the inhabitants of the area in the future. One of our respondents put it as follows: 'we have to complete boxes that literally ask, how are we addressing these issues? What are the implications for the future? No, I do think it's important that, definitely'.
Recent statements have also focussed on the Welsh Government's capacity to deliver these more hopeful futures, particularly post-Brexit. The Welsh Government's (2017a: 24) position is now that 'we want to take the best elements from our EU legacy and build a new future for the long-term'. In the same document, emphasis is placed on Wales' capacity to influence its own future, especially since leaving the EU risks reinscribing UK-level macroeconomic policy on the devolved Welsh Government. This involves a call for 'shared sovereignty' with future ' UK Government investments requir[ing] closer coordination with the Welsh Government to ensure they work with, not against, the grain of Wales' economic policy' (Welsh Government, 2017a: 20). Furthermore, the Welsh Government (2017b: 27) has called for new forms of collaboration between Wales and the broader UK post-Brexit to protect Wales' capacity to influence its own economic future; a capacity understood in relation to politics and sovereignty, and the ability to shape distinctive and relevant policies for Wales.
While we might decry the perhaps unrealistic ambition for Wales to be the 'best place to live, learn, work and do business', there is a clear attempt to envision a future in which Wales will be more prosperous, sustainable and just. This marks a discursive shift from the self-portrayal of Wales as the failing victim of processes beyond its control towards a future that Wales can itself influence. We see, in short, the beginning of a vision that seeks a more spatially just future for Wales.

-New spaces and scales of governance
Our second empirical theme is concerned with the way in which territorial cohesion has influenced spaces of governance at the regional scale, and how Brexit and spatial justice signal the potential for new spatial transformations. Problematic spatialities are being rightly questioned, and new spaces and scales of regional governance are emerging; with the hope that these emergent forms will prove more spatially just.
The effective operationalization of EU cohesion policy through economic indicators (which we discuss in the following section) has led to the literal mapping of NUTS2 regions into 'less developed,' 'transitional' and 'more developed' categories (see Figure 1). Of the two NUTS2 regions in Wales, East Wales is classified as 'more developed,' with a GDP per capita greater than 90% of the EU average. The predominantly rural and post-industrial West Wales and the Valleys region has a GDP per capita below 75% of the EU average, and is classified as 'less developed' (formerly known as 'Objective 1'). Yet, several scholars have drawn attention to how NUTS2 boundaries were deliberately redrawn in Wales in order to maximize EU funding eligibility (e.g. Boland, 2005;Gripaios and McVittie, 2003). A 1999 document produced under the auspices of Institute of Welsh Affairs put 'the case for redrawing the deprivation map of Wales from the old north/south division to a new east/west configuration' (Osmond, in Morgan and Price, 1999: 2), and noted Eurostat's assent to a corresponding NUTS2 change. The report's authors justified the new division: For ordinary men and women, Wales today stands as a country clearly divided between the relatively prosperous, confident and accessible areas of the east, and the de-industrialised or crisis-torn rural communities to the west (Morgan and Price, 1999: 7).
However, Gripaios and Bishop (2006: 939) observed just a few years later that the redrawn NUTS2 regions 'seem to lack economic logic and appear to be contrived to ensure that as many of the 22 Unitary Authorities in [Wales] as possible qualify for [EU] funding'. The contrived separation of an allegedly prosperous 'East Wales' from a deprived 'West Wales and the Valleys' was readily acknowledged by many of our interviewees. As one straightforwardly explained: In order to achieve eligibility for European funding … Wales was sort of cut in half. So, you've got what they call West Wales and the Valleys, which are the bits that are eligible for the highest level of European funding. Then you've got the rest … all the good bits economically (INT03, senior advisor, economy).
Multiple consequences have arisen from this spatial division. First, as we maintained in 4.1 above, the division has helped sustain a narrative of Wales, and particularly of post-industrial and rural spaces, as failing and dependent. Second, contrived NUTS2 boundaries have been imposed on other, more meaningful, regions and localities. Neither East Wales nor West Wales and the Valleys are coherent regions functionally, geographically or historically. Further, funding boundaries have created an artificial dislocation between contiguous areas and pre-existing functional regions. For instance, an important functional region extends from the relatively prosperous Welsh capital, Cardiff, to some of the most deprived local authorities in the South Wales valleys; these connections are severed by the artificial assignment of these places to separate NUTS2 regions. Several interviewees drew attention to the political and practical difficulties such separations induce. For example, an advisor working with local government explained the difficulties of developing coherent funded programmes and interventions that crossed NUTS2 boundaries. Two separate applications would be required for the same project (one for each region) and different levels of funding would be given to each, and would need to be meticulously accounted for. No matter how much practical sense working across the NUTS2 boundaries makes, interventions are often administratively stymied.
These working difficulties have also been acknowledged by the Welsh Government's white paper Regional Investment in Wales After Brexit, which states: The division of Wales into regions that reflected the primacy of EU statistical classification in determining eligibility for, and hence management of, EU funding streams, was consistent with the EU-wide approach to regional allocation, but did not in all cases align with local level need (Welsh Government, 2017a: 18).
The UK's imminent departure from the EU has indeed been associated with some relief that regional investment policies can now be rethought. The white paper anticipates new geographical possibilities for regional development beyond the NUTS2 map: A potential benefit of being outside of the EU will be the opportunity to work more systematically with functional regional areas that reflect the economic reality in each part of Wales, rather than being constrained by the current geographical and fund-specific limitations. We will no longer need to separate parts of West and East Wales artificially, or address the needs of rural areas, people, and businesses entirely separately (Welsh Government, 2017a: 14).
Alongside reiterating the challenges of promoting economic development and social policy through artificial, instrumentally-defined regions, this statement interestingly highlights the possibility of working with 'functional regional areas that reflect the economic reality in each part of Wales'. Some new and potentially innovative spatial imaginaries are at work here, including interconnections between cities and rural hinterlands, and productive links between North Wales and the adjacent 'Northern Powerhouse' in North-West England (Welsh Government, 2017a: 19). Recent emphasis on city regions as a new scale of governance within the UK (Harrison, 2012) provides an additional emergent spatial reference, with the Welsh Government's post-Brexit vision including four regional 'growth deals'. There is already some evidence that the growth deal model has influenced how the current round of EU structural funds have been used in Wales. For instance, a local government interviewee described how recent projects were responding to the need to develop 'more regional ways of working' by trying to attend more to city regions than NUTS2 agglomerates.
There are, therefore, interesting attempts afoot to envisage different geographies of governance for a post-Brexit Wales--geographies unconstrained by the instrumental responses to, and regulatory requirements, of territorial cohesion. Our interviews and documentary research indicate a growing admission by Welsh policymakers that the regional geographies associated with cohesion have been meaningless, unhelpful and, by extension, unjust. Yet, we might ask whether emergent regional approaches are any more grounded in notions of spatial justice. While city regions and growth deals do place some emphasis on accessing services and addressing perceived social need, the policy documents we consulted largely view these approaches as economic ends (e.g. Welsh Government, 2017a). There is thus a danger that in sticking to a narrow focus on certain indicators of economic success, new regional ways of working could simply replicate the problems we have identified with the existing NUTS2 boundaries.
New spaces and scales of governance have also appeared in the context of the Well-being of Future Generations Act. Although not explicitly spatial in approach, the Act is grounded in localities in Wales through the creation of Public Services Boards (hereafter PSBs), which are largely based on local authority boundaries, and which are tasked with producing local well-being assessments and setting localized objectives towards the national well-being goals (Jones, 2019). The first tranche of assessments were released in 2017. This delegation of responsibility to PSBs can be viewed as a positive attempt to spatialize justice within Wales by enabling different understandings of well-being to emerge in different local authority areas. Different emphases have indeed emerged on the ground. The well-being assessment from post-industrial Torfaen, in south-east Wales, for example, highlights the challenges associated with chronic health conditions, inter-generational patterns of poverty, and the need to provide a healthy start for children and young people (Torfaen PSB, 2017). By contrast, Gwynedd and Anglesey PSB (2017), in rural and Welsh-speaking north-west Wales, has identified the need to preserve communities, protect and promote the Welsh language, understand the effects of demographic change, and ensure the availability of affordable housing (Jones and Lewis, 2019).
The above evidence demonstrates that the impact of the Well-being Act extends well beyond a shift in national rhetoric, important though that is. The Act has led to the emergence of new well-being priorities in different parts of Wales. Under previous understandings of territorial cohesion, Gwynedd and Anglesey and Torfaen would have been subsumed into West Wales and the Valleys and a common socio-economic indicator relating to GDP would have been applicable in both areas. Now, with a new focus on more localized interpretations of spatial justice, alternative visions of wellbeing can flourish. As a respondent from Gwynedd put it simply, 'there's not going to be the same solutions for everywhere, is there?'.
Nevertheless, questions arise about the extent to which local well-being assessments really reflect a spatial approach to well-being. In a publicly released review of Well-being Assessments and Goals, the Future Generations Commissioner for Wales (established under the Act) commented that: The assessments acknowledge the importance of local spaces as assets, acknowledge their roles in people's well-being and consider how people engage and interact with these places. However, most well-being assessments showed very limited consideration of the significance or cause of spatial differences (OFGC, 2017: 15).
The Commissioner's feedback encourages PSBs to think more deeply about the relationship between space and well-being; to move beyond viewing space as merely a container for (in)justice, viewing it instead as something that contributes to (in)justice in far-reaching ways (Dabinett, 2011(Dabinett, : 2391.

-
Beyond GDP: imagining new measures of success A third shift in discourses revolves around the definition of success and the 'good life'. We discern growing efforts in Wales to broaden conceptualizations of success beyond macroeconomic measures like GDP/GVA, instead taking into account broader conceptualizations of well-being and justice.
Many of our interviewees critiqued what they saw as an over-emphasis within territorial cohesion policies on GDP as a measure of regional success or failure. Despite discourses of 'investment', territorial cohesion is still measured according to a normative economic view in which the ultimate concern is the regional distribution of GDP. Yet, as we considered in the section above, many interviewees were highly aware that what gets counted as a regional inequality depends on where the dividing lines are drawn. For them, economic indicators particularly concretized the contingent in unhelpful ways. GDP was especially criticized as a 'blunt instrument … [that] doesn't really tell us much about what's going on underneath' (civil servant, Welsh European Funding Office). Similarly, a senior advisor compared economic indicators to well-being, reflecting: nowhere do we say the purpose of our economic policy is to increase GVA. Whereas the comparisons that we tend to do nationally, internationally are [on this basis] … and that masks a whole range of other things … Your GVA could be going up and that can still mean that you've got some terrible things happening for particular types of people and particular communities (senior advisor, well-being).
This comment suggests how exclusively economic measures of regional success can fail to connect meaningfully with local development needs. The dangers of this mismatch are self-evident, and include, as we described above, attempts by regions to redefine how they are constructed in EU statistics.
Issues of comparison are a further consideration. While common measures like GDP offer scope for useful comparisons across the EU map, they can also lead to unhelpful comparisons between regions. For example, some interviewees perceived GDP as a tool used to critique regions deemed to be consistently underperforming: 'we have been pointed towards regions in Europe that had started off as receiving the highest levels of aid and then have gone to grow out of those … So, they've done it. Why can't you?' (civil servant, Wales European Funding Office). While territorial cohesion compares Wales statistically to purportedly more successful regions, Welsh policymakers reported that they were more interested in more meaningful comparisons with regions of similar size and circumstance. Largely, this involves 'using the best of other countries' (senior civil servant, social services) through good practice models, rather than benchmarking indicators. The majority of the individuals we interviewed resisted Wales being 'measured up' with other regions, and were neither intellectually nor professional engaged by notions of 'catching up' with them. Rather, they wanted to tackle inequalities within Wales.
By ascribing success in macroeconomic terms and through regional comparisons, therefore, territorial cohesion is far from enabling. As Lea (2008: 16) insists, through such an 'interventionary lens', 'the idea that life might be lived differently with value and meaning or that 'need' might be conceived of differently from the way in which we calculate it … becomes impossible to imagine'. A lens focussed on macroeconomic indicators makes it especially difficult to imagine or enable more regionally appropriate definitions of well-being and the good life. Spatial justice, we contend, can provide an alternative route for regions to actively define their own futures, using their own measures of success (Sen, 2009).
We hinted above at how the Well-being of Future Generations Act is seeking to promote alternative approaches to development and the good life. The 'Wales We Want' consultation exercise aimed to develop a distinctly Welsh take on sustainable development, reflecting Welsh values (Jones and Ross, 2016). This included a concerted effort to highlight notions of justice. A Sustainable Development Report published by the Welsh Government (2012: 5) in 2012 stated that Wales 'has its own account of sustainable development', with an 'emphasis on social, economic and environmental wellbeing for people and communities, embodying our values of fairness and social justice'. The Welsh Government's (2014: 1) submission to 'The Wales We Want' similarly stated that '[w]e need people to be healthy, to achieve their potential and to make Wales a more equal society'. Interviewees who engaged with civil servants in other jurisdictions further emphasized Wales' distinctive approach. As a senior manager from Natural Resources Wales (the organization responsible for managing the Welsh environment) explained: If you look at our aim [Natural Resources Wales], the remit letter we receive from government talks about outcomes and it's true that that relates to more than just the environment. There's far more of an emphasis on reducing poverty etc than you would find with the Environment Agency, with the Countryside Commission in England.
This emphasis on notions of justice within Wales' interpretation of sustainable development was reinforced by the renaming of the Sustainable Development Bill as the Well-being of Future Generations Act. Interview evidence suggests that part of the reasoning behind this change in nomenclature revolved around a perceived need to highlight the broader understandings of well-being and justice that were to be promoted by the Act (Jones and Ross, 2016).
The Act arguably, therefore, possesses a distinctly Welsh 'take' on sustainable development, well-being and justice. The Welsh Government's commitment to achieving these just futures is made concrete in the form of the various measures of progress towards future well-being. Rather than the macroeconomic 'blunt instruments' our interviewees critiqued above, the Act experiments with different ways to assess wellbeing. There are currently 46 indicators tracked by Statistics Wales as part of the Act, including: the gender pay gap (indicator 17); percentages of people moderately or very satisfied with their jobs (indicator 20); satisfaction with access to facilities and services (indicator 24); the proportion of the population that can speak Welsh (indicator 37) (Welsh Government, 2015b;2015c). It is these multiple measures, above all else, which testify to Wales' attempt to recalibrate understandings of the just futures that the Welsh government should be attempting to create.
Admittedly, these alternative measures can be criticized. The Welsh Government's (2015c) technical guidance especially leaves us with a sense that the indicators chosen are less innovative than reflective of data that is already being collected by different government agencies, such as the UK Labour Force Survey and the National Survey of Wales. There is a suspicion, accordingly, that other potential vectors of spatial justice might well have been omitted, partly because the collection of data to monitor progress in relation to them had been viewed as being either difficult or costly. Yet, new questions are evidently being asked by policymakers: 'how good are our relationships? What's the area we live in like? … Can we travel to see friends and do activities and do all of those sorts of things?' (senior civil servant, well-being). For many working on the well-being agenda, these kinds of questions do represent a new departure. And as we showed earlier, policymakers working in different various local authorities and linked to different PSBs are beginning: (1) to ask different questions about their localities; (2) emphasize contrasting visions of well-being (in assessments and future goals). We contend that the attempt to develop augmented understandings of well-being and, by extension, justice, point the way to how an approach informed by spatial justice principles might allow policymakers to think more creatively about the kinds of regions they are trying to create.

Conclusions
Although broader in premise and potential, the EU's concept of territorial cohesion has materially proceeded through a concern with distributions of financial resources and narrowly defined macroeconomic measures of developmental 'success'. In Wales, the heavy financing of regional development has reflected both persistent inequalities and what we have described as a reactive narrative of victimhood and dependence. After 18 years of structural funds investment, West Wales and the Valleys is still classified as a 'less developed' region, and discourses of failure continue to be affixed to Wales as a whole. Unsurprisingly, our interviewees considered that European funding has been ineffective in addressing regional inequalities in Wales. We agree; yet not just on the basis of stalled indicators or deprived communities. As we have argued, the operationalization of territorial cohesion goals has failed to enable alternative, more regionally apt visions of well-being and the good life to emerge. Through the macroeconomic map imposed by territorial cohesion policies, Wales can only ape more successful European regions, using rules those 'winners' have already set by default.
But the evidence from Wales also suggests that other approaches to addressing inequalities are possible. We discern a notable turn in Welsh policy away from narratives of 'Welsh problems' towards envisioning, to borrow the title of the Welsh Government's recent sustainable development consultation exercise, 'The Wales We Want'. As we have argued, these changes constitute a shift in regional development approaches that resonate with--and have the potential to be more directly informed by--principles of spatial justice. Three key shifts were highlighted above: a change from narratives of dependence to regional capacities to act; the emergence of new spaces and scales of governance; a redefinition of measures of success beyond macroeconomic indicators. While we neither contend that these shifts embody spatial justice, nor that they are the inverse of the territorial cohesion agenda, we do believe that they constitute positive steps for Welsh public policy. More broadly, we consider that these changes suggest possibilities for regional development approaches elsewhere that are directly informed by spatial justice. This is not to suggest a transferable 'Welsh model'. Rather, the case of Wales suggests that regionalizing spatial justice is possible, preferable and, amidst the crisis of EU Cohesion Policy, necessary.
The discussion also poses some significant conceptual questions about spatial justice. With a few notable exceptions, much of the research on spatial justice to date has concentrated on the city as a space and scale. The case study of Wales shows, however, that academics and policymakers must grapple with some serious issues when attempting to 'translate' spatial justice to the regional scale (Prince, 2012). At the very least, spatial justice calls for approaches to be tailored to each specific region's distinctive visions and capacities to act. Other questions abound. To what extent can one pursue spatial justice solely at a regional scale, thereby underplaying the specific needs of particular localities? To what extent does envisioning justice at the larger scale of the region change the spatial, socio-economic and political dynamics through which that justice is achieved? To what extent does defining visions of well-being and justice at a regional scale, unless one is careful, lead to unhelpful and unrealistic boundaries being created between contiguous regions? To what extent would a pursuit of spatial justice at the regional scale inevitably privilege metropolitan and urban interests at the expense of rural ones (cf. Jones and MacLeod, 1999)? Conversely, can a delegation of the definition of spatial justice to more local scales within regions--as has been the case with PSBs in Wales--lead to a situation in which understandings of justice vary geographically in potentially unhelpful ways (e.g. in relation to access to services)? In short, we need to examine the extent to which the spaces and scales over which justice is being articulated influence the extent to which justice can be realized.
If these are important issues in the particular context of Wales, they are even more pertinent when one seeks to upscale the idea of spatial justice to Europe as a whole. One important thread within the literature on spatial justice, as we have seen, is that individuals, groups, cities, regions and states can seek to define their specific understanding of justice (Storper, 2011: 19). If this is the case, then equal scope must be given to those same stakeholders to articulate the spaces and scales over which that search for justice occurs. The upshot of all this is that the most appropriate spaces and scales to achieve justice might vary from one place to another; spaces and scales of more just governance might work more effectively in some places than others. There is a need, therefore, for a spatial sensitivity in relation to spatial justice. Social scientists should lead in attempts, not just to analyse this spatial variation, but also to enable the most effective governmental configurations to be developed to promote spatial justice within particular geographical settings.