Abstract

Cities and their hinterlands have always had economic, social and cultural links.  However, the governance arrangements vary enormously, from fully integrated administrations to separate and often competing political units.  Today, with Britain’s cities widely viewed as the main motors of regional economic growth, there are moves towards new structures of city region governance across the UK.  These involve increased coordination between existing government bodies and interest groupings, though there are marked geographical variations in progress towards such collaboration.  This paper considers whether these variations in governance structures really matter.  It is based on new research into the relationship between the UK coalfields and their neighbouring cities.  The coalfields are one of the principal areas of industrial restructuring in the UK, having lost virtually all the jobs in the coal industry itself since the early 1980s.  In theory, neighbouring cities might offer economic salvation for such areas, which could develop a new future as residential and ancillary service hinterlands.  But is this really the emerging trajectory, or are the former coalfields experiencing a revival independent of their neighbouring cities?  And do governance structures make a difference to what is happening?  Research evidence from three coalfields (Lothian, South Yorkshire and South Wales) and their neighbouring cities (Edinburgh, Sheffield and Cardiff) reveals that collaborative governance structures spanning these areas vary markedly, as do the economic trends.  Overall there appears to be no close relationship between the existence of collaborative governance and the strength of coalfield-city economic links.  On the contrary, cross-boundary collaboration appears to have greatest relevance in reconciling the competing claims of rival localities and hence in fostering polycentric development, rather than in promoting city-focused growth.  Key factors driving area regeneration are underlying economic strengths and weaknesses, the operation of the market, the availability of funding, and the suite of policy tools through which it is allocated.

Introduction

As the UK has moved further away from its position as one of the world’s industrial powerhouses, people and places have had to adapt to new social and economic requirements.  Increasingly open local economies raise questions about how growing and complex networks of links between places alter the role and identity of former industrial areas.  These questions relate closely to key contemporary issues including geographical disparities in welfare and productivity, the uneven operation of the labour market, the governance of economic development, and the competing claims of ‘regions’, ‘sub-regions’ and ‘city regions’ as the appropriate arena for sub-national policy-making.  Indeed, within the UK there is currently a lively policy debate surrounding the most appropriate strategic scale for sub-national economic development, and the extent to which different types of place should be prioritised within their spatial frameworks.

This paper is intended as a contribution to this debate.  It presents a comparative, empirically based examination of the extent to which the emergence and growth of city region collaborative governance is actually associated with greater economic integration between city and hinterland.  It does this by drawing on a recent study of three former British coalfield areas and their neighbouring cities (Lothian-Edinburgh, Central Valleys-Cardiff, and South Yorkshire-Sheffield) (Gore et al., 2007).  The study investigated economic, social and governance links between the coalfields and their cities, and assessed whether the coalfields are reviving in their own right, or are more likely to become residential areas and lower-level service providers for newly resurgent cities.

Table 1 summarises the relative sizes and labour market positions of these three study areas.  There are clear differences between them in terms of both geographical extent and population size, as well as between each city and its coalfield hinterland.  Including this range of places was a deliberate choice, the aim being to investigate the influence of different contextual factors on patterns of integration and cross-boundary collaboration.  It should also be acknowledged that past historical links between the three cities and the coalfield areas were quite varied as well.  Thus, apart from its erstwhile port functions at Leith, Edinburgh’s role in the development and exploitation of Lothian coal was quite limited.  However, its strength as an employment centre and the relative proximity of the two areas has meant that the coalfield always generated some city-bound commuting; as will be seen, this has now grown substantially due to Edinburgh’s continued employment growth and the associated extensive new house-building in the former coalfield.

In contrast, links between Cardiff and the Central Valleys actually dwindled with the demise of the coal industry.  Indeed, the original prosperity of the capital city of Wales stemmed from its role as entrepot for Valleys’ coal exports, a position that may underlie some of the current antagonism between them (the city profiting at the coalfield’s expense).  More recently, the growth of Cardiff as a commercial, financial and administrative centre has been built upon links eastwards towards England, Europe and the world, rather than northwards to the Valleys.  However, the strength of its economic position means that it remains the dominant force as far as the Central Valleys are concerned, compounded by the long-standing economic fragility of the coalfield area itself.

Sheffield and the South Yorkshire coalfield present yet another different story.  Here, apart from the use of locally produced coal in the city’s steel furnaces, for years there were few if any links between them.  The market for – and indeed reputation of – the specialised steels produced was international, rather than local or regional.  At the same time, Sheffield never developed an extensive commercial class on the back of manufacturing industries in its hinterland, in the manner of Leeds or Manchester.  In many ways its very specialisation made it rather introverted, at least in relation to its surrounding area, if not in terms of its global trading links.  The loss of the steel industry’s dominance some 25 years ago has left the city searching for a modern-day role.  While it remains in the shadows of stronger regional centres such as Leeds, Manchester and Nottingham (all just around 40 miles away), its lack of strong linkages to its immediate hinterland also means that it struggles to assert any kind of sub-regional authority.

The paper is divided into five main sections.  The first sets out the main parameters of the debate around the appropriate scale for sub-national economic policy in the UK.  The second summarises recent trends in commuting between the three coalfields and their nearby cities.  The third examines the job search patterns of unemployed coalfield residents, with particular reference to the place of the neighbouring city in these, and the role of respondents’ social networks.  The fourth reviews the extent of collaborative governance arrangements across the three areas; and the fifth draws together conclusions and recommendations.

Table 1: Profile of the Three Study Areas

				Employment Rates
	Area (sq. km.)	Total Population	Working Age Population	Males	Females
Central Valleys coalfield	755	435,000	265,000	65.8	59.0
Cardiff County Borough (local authority area)	140	315,000	205,000	73.9	67.6
South Yorkshire coalfield	1004	655,000	405,000	70.9	64.5
Sheffield City Council (local authority area)	367	515,000	320,000	77.0	67.8
Lothian coalfield	457	120,000	65,000	79.8	75.4
City of Edinburgh Council (LA area)	262	450,000	295,000	77.5	72.1

Sources: Regional Trends (2001); Mid-year Population Estimates 2003; Local Area Labour Force Survey, 2003.

Note: Population estimates have been rounded to the nearest 5,000.

City Regions and territorial rescaling

Since the devolution of selected powers to Scotland and Wales in 1997, there has been a lively debate within the UK (and particularly in England) about the most suitable strategic scale for sub-national economic development policy.  A key element of this has been the degree of importance to be attached to different places, in terms of the potential economic benefits they might generate, and the nature of the spatial policy framework that would foster these most effectively.  Although there are shades of opinion, at the political level the protagonists now broadly fall into three camps.

First, there are those that argue that the interdependence between cities and their hinterlands is of fundamental importance in terms of regional economic development, productivity growth and competitive performance.  As the third report of the UK Government’s Core Cities Working Group, Our Cities Are Back claims: ‘if you can improve the economic performance of cities this will have a major impact upon the economy of the entire region’ (Core Cities Working Group, 2004: 26).  In this model the dynamism of the city and its role as a hub in the emerging ‘knowledge economy’ produces ‘spillover’ effects within the surrounding area.  These arise especially in terms of provision of ancillary goods and services, access to higher quality jobs, a wider range of urban and rural housing options, development of niche markets and increased leisure activities (ODPM, 2003).  As a result, many parts of Britain are actively pursuing city region arrangements that focus on the cities acting as ‘drivers’ for the wider area.  An influential voice in this process has been the Work Foundation on the city region (or ‘Ideopolis’) as the most effective spatial unit for the wider spread of the knowledge economy (Jones et al., 2006).

A second group has been sharply critical of these arguments in favour of ‘monocentric’ city development, seeing them as simplistic and unproven, especially with regard to the mechanisms whereby benefits can be spread to wider metropolitan and rural hinterlands.  A good example of this counter stance is a recent report issued by the Chief Economic Development Officers Society (CEDOS) and the County Surveyors Society (CSS) (2007), which argued for a much more inclusive approach.  “….analysis shows that all areas….have distinctive roles to play as part of a complex mosaic that forms the modern economic geography of Britain…In reality, regions depend on the inter-relationships of a series of complementary and competing areas, with ‘boundaries’ varying and indeed overlapping according to the particular mix of definitional factors being used.”  (ibid: 5)  This points instead to an intricate polyzonal pattern of multi-layered, multi-level jurisdictions through which all group and area interests might be represented.

The third influential voice has tended to navigate a middling compromise course between these opposing factions, emphasising the need for continuity, clarity and local acceptability, and is closely associated with the views of prominent representatives of HM Treasury.  The collection of papers written by Balls et al. (2006) illustrates this stance well.  The gist of their case is that it would be wrong to assume that the successful London model, based on its unique position in the UK as a ‘world city’, can simply be transposed to all other conurbations.  These have very different structures, histories and mixtures of local identities.  These authors contend that the reallocation of roles and responsibilities to new city region arrangements could well destabilise the existing multi-level approach that brings together the RDAs, sub-regional partnerships and local authorities and other partner agencies.  Instead, they consider that a combination of strong regional co-ordination and increased local flexibility, aligned to the specific conditions and circumstances of particular places, would be the best way to increase prosperity and reduce inequalities.  In many ways this approach defers to the arguments of both camps, with wider collaboration such as at city region scale encouraged on a voluntary basis, but no part of the country excluded from the system as a whole.  This ethos can certainly be detected underpinning the revised sub-national governance arrangements for economic development and regeneration proposed in the Government‘s recently published review (HM Treasury, DBERR and CLG, 2007).

Lying behind these exchanges are the changing nature of Britain’s economic geography and the associated importance of attracting globally mobile capital investment and the job opportunities that go with it.  Highly differentiated manufacturing and service operations now serve disparate (and often dispersed) markets, and this brings with it a plethora of locational, infrastructure, business support and labour supply requirements (Taylor et al., 2006).  Any given locality is likely to find it difficult to meet this wide range of needs, and this has increasingly prompted a quest for co-ordination of policy interventions across larger spatial units.  During the late 1990s this was pitched particularly at the regional scale, with political devolution to the UK’s non-English ‘nations’, and administrative devolution within England.  However, the crisis suffered by the English regional devolution project following the ‘No’ vote in the North-East referendum in 2004 offered a boost to the emergent city region concept as a more appropriate spatial framework for sub-national policy-making.

At the same time, in academic as in policy-making circles there are competing visions of how city regions work.  In broad terms there are three models.  First, a single central city serves its surrounding, largely rural hinterland (the “monocentric” or “core-periphery” model).  Second, some urbanised areas have two main centres or zones of development that link into the intervening area (the “bi-polar” model).  Third, there are broader conurbations comprising multiple service centres and employment zones (the “polycentric” model) (Herrschel and Newman, 2002).

In broader terms, the turn to city regions can also be interpreted as an example of territorial rescaling by the state.  According to this, the fixed spatial frameworks embedded in existing institutional arrangements no longer even approximately match the increasingly complex spatial networks that characterise capitalist economic processes and reference points for locational decision-making (Dicken, 2003).  Hence, according to writers such as Brenner (1999; 2004) and Jessop (2002) there is a search for a more flexible and responsive territorial articulation of state institutions and activities.  This at the same time has to contend with high levels of resilience on the part of existing organisational forms at the sub-national level.  One way of overcoming such resistance, of course, is to try to incorporate such bodies into wider processes of cross-boundary collaboration, for example at a city region scale, while simultaneously preserving their general integrity.

Clearly, the move to city regions in the UK is about more than just regulation or promotion of economic development.  As Morgan (2006) points out, it is also inextricably tied up with the critical need for democratic and political legitimacy for the English regional devolution project, while at the same time promising to make a major contribution to securing greater efficiency in public service delivery.  He also highlights the inherently political nature of the city region approach, identifying the key players behind its promotion and their vested interest in doing so.  Using Cardiff and the South Wales Valleys as a case study, he emphasises that such a partial approach to collaboration is likely to be unsuccessful.  As in all political projects, there need to be rewards on offer for all those taking part, and as yet these have been slow to emerge for those organisations responsible for governance in the Valleys themselves.

However, one feature that does not seem to be in doubt is the veracity of the city region idea itself.  In the words of Leslie (2006), “everyone recognises the realities of ‘city regions’ as economic entities….” (p.9). Here the thinking of political commentators and academic analysts starts to diverge.  Thus, the literature on territorial rescaling suggests that the complex and overlapping spatial networks characterising contemporary economic activity are not reducible to a simple geographical construct like a city region.  The likely result would merely be to replace one ‘mismatched’ set of institutions with another.  What the rescaling approach does emphasise, however, is the essential importance of co-ordination and collaboration over broader geographical scales.  However, there is a danger that the shapes and structures of rescaled governance become viewed in an overly prescriptive, even normative manner.  In fact, previous experience indicates that the exact coverage and nature of these arrangements generally have to develop by trial and error, with varying responses to particular local circumstances and political realities (Gualini, 2006).

Previous research on city regions and territorial rescaling has tended to focus on places where such developments have progressed or are developing to a significant extent.  However, this begs the question why such trends have not been universal throughout a particular country such as the UK, if the economic processes that lie behind it are operating at a global scale.  In line with the research agenda outlined by Gualini (2006), this calls for further comparative research into the contexts and circumstances in which moves towards city region working in particular and collaborative governance in general do or do not manage to take root.

Labour market connections

Across the British coalfields as a whole, people moving out of the area or starting to commute to places elsewhere have partially compensated for job losses in the coal industry, but they account for only a relatively small proportion of the overall adjustment.  The dominant trends have been within the coalfield labour market itself, particularly new job creation (on the positive side) and withdrawal of men into economic inactivity (on the negative side) (see Table 2).

Two lines in Table 2 say something about how the coalfields relate to other areas, including neighbouring cities.  The first concerns migration.  The negative figure for ‘net in-migration’ for men indicates that the coalfields did on balance lose men of working age to other areas.  This is to be expected in areas of job loss, but is counterbalanced by the fact that it is far exceeded by other headings in the accounts (particularly jobs growth in the coalfields as well as the aforementioned increase in economic inactivity).

Table 2: Labour market accounts for the English and Welsh coalfields, 1981-2004

		Men		Women
		No	as % working age in 1981	No	as % working age in 1981
	Job loss in coal	222,000	15.9	–	–
PLUS	Natural increase in workforce	86,600	6.2	60,200	4.8
PLUS	Net in-migration	-57,900	-4.1	3,300	0.3
PLUS	Increase in net in-commuting	-30,700	-2.2	-72,500	-5.7
PLUS	Increase in economically active	-162,500	-11.6	125,000	9.9
MINUS	Increase in non-coal jobs*	132,400	9.5	131,500	10.4
EQUALS	Increase in recorded unemployment	-74,900	-5.3	-15,800	-1.2
*For women, includes job losses in coalSources: Census of Population, Annual Business Inquiry, ONS, Coal Authority

The second relates to commuting.  Here the figures show that although the net outflow from the coalfields has increased, for men this has been a minor part of the overall adjustment process, involving just 2 per cent of the working age male population.  For women the rise in net out-commuting has been somewhat more important (covering 6 per cent of the working age female population), although again far more have moved into jobs within the coalfields themselves.

The scale of these new jobs has partly depended on location and context, but, more importantly, they have arisen in the coalfields themselves, rather than in neighbouring areas such as the cities to which coalfield residents might commute.  In other words, the success or failure of economic regeneration in the coalfields as a whole has principally depended on trends in the coalfield economy itself, not in neighbouring areas.

However, this is not the case for every coalfield (see also Beatty et al., 2005).  The three areas studied each display different relationships with their neighbouring cities:

• the South Yorkshire coalfield is a large, relatively self-contained economic zone, whose economic fortunes have largely been determined by what has happened within its own boundaries.  Job creation has been strong in recent years, and there is no evidence of any huge increase in commuting to neighbouring areas (see Tables 3 and 4).  Links with Sheffield do exist, of course, but they are not driving overall change in the coalfield labour market.  The increase in female commuting looks impressive, but it is in part affected by starting from a fairly small base.  Transport access to Sheffield is also difficult from many parts of the coalfield.  Within the coalfield, established urban centres such as Barnsley, Doncaster and Rotherham act as more important hubs than Sheffield, as do newer employment zones such as the M18 corridor, the Dearne Valley and other reclaimed colliery sites.  Here there has been substantial growth in male employment in construction, distribution and logistics and telecommunications, whilst the brunt of recent job losses in manufacturing has affected women
• the Central Valleys coalfield in South Wales also remains a substantial economic zone in its own right, but new job creation has been relatively weak (see Table 4).  In this context, Cardiff’s exceptionally strong employment growth in recent years has brought a rise in commuters from the coalfield.  The Central Valleys have witnessed an increase in commuting links with their neighbouring city, although the proportion of working-age residents travelling to jobs in Cardiff remains relatively small at just under 10 per cent (see Table 3).  The growth in commuting is more a symptom of economic weakness in the Central Valleys themselves.  There is also a contrast in commuting volumes between the more accessible southern fringes and the more distant northern parts of the coalfield
• the links between the Lothian coalfield and Edinburgh are very strong and continuing to grow, based on their fairly close proximity and good transport links.  Along with its relatively small size, this means that the coalfield is less an economic zone in its own right and more part of the city’s interdependent hinterland.  Commuting into Edinburgh is a dominant feature of the Lothian coalfield (see Table 3), and this has been rising through time.  Population has also been rising, mainly through new housing developments that cater for in-migrant Edinburgh commuters.

Table 3: Commuting, 2001

	% of working age-population			% change 1991-2001
	Men	Women	All	Men	Women	All
Central Valleys coalfield to Cardiff	8.9	8.0	8.5	+23.5	+30.7	+26.7
South Yorkshire coalfield to Sheffield	7.3	5.7	6.6	+6.0	+26.8	+13.9
Lothian coalfield to Edinburgh	37.8	41.2	39.5	+20.7	+42.7	+31.2

Source: Census of Population, Special Workplace Statistics, 1991 and 2001

Table 4: Employment growth, 1998-2003

	% change
	Men	Women	All
Central Valleys coalfield	0.0	4.0	2.0
Cardiff	17.2	19.5	18.3
South Yorkshire coalfield	11.1	8.0	9.5
Sheffield	6.3	7.9	7.2
Lothian coalfield	7.5	8.3	7.8
Edinburgh	7.6	16.3	12.0

Source: Annual Business Inquiry (via NOMIS), 1998-2003

Job search patterns, social networks and coalfield-city links

The study also involved interviews with 85 unemployed job-seekers living in the three coalfield areas.  These provided information on geographical aspects of their job search, such as attitudes to commuting and general mobility, and on their local attachments and social networks.

Respondents’ willingness to consider working in the adjacent city only partly mirrors the differences in labour market trends detected by the quantitative analysis.  Thus, as expected a majority of Lothian respondents were searching for jobs in Edinburgh, partly because this fell within their definition of the ‘local area’.  In contrast, South Yorkshire occupies a ‘middling’ position – more people than expected had a positive attitude to working in Sheffield, depending on the exact location and type of work.  Far fewer respondents in the Central Valleys included Cardiff in their job search than the recent increase in commuting might suggest.  However, these interviews included several young people (aged between 17 and 22) living further from the city in the north of the Valleys area, which will have affected the overall pattern of the responses.  As previous research by Quinn (1986) and Green et al. (2005) has illustrated, young people tend to have more limited spatial experiences and horizons, and hence are less likely to consider job opportunities in more distant and unfamiliar locations.

The interviews revealed that travel to the neighbouring city for work is constrained by a number of factors.  For many, simple economics generally ruled it out – commuting costs would swallow up a large proportion of a typical wage.  The time required for the return journey was a deterrent for many, while others were restricted by the lack of early-morning or late-evening public transport services, or by childcare responsibilities.

Knowledge and experience of the city were also important factors.  Those who did include it in their job search generally focused solely on areas they know, such as the city centre, or along the route between there and their home.  This effectively excluded large areas of the city from consideration.  Moreover, not many had access to private transport, so were reliant on public transport to get them to work, limiting them to workplaces near bus routes or railway stations, and to working hours that coincide with service operating times.

Most respondents had closed, locally based social networks built around family and friends.  These networks provided important sources of support and information for job search.  However, the support appeared to be more important than the information, as most used conventional sources when looking for jobs, such as local newspapers, the Job Centre or employment intermediaries.

However, the local nature of people’s social networks did play an important role in influencing attitudes to acceptable types and locations of work.  In particular, it meant that many respondents had been exposed to limited experience and knowledge of other areas.  There was also a contrast between the Lothian and South Yorkshire coalfields on the one hand, and the Central Valleys on the other, with respondents in the Valleys area more likely to have just unemployed or inactive people in their social network.  In Lothian and South Yorkshire most respondents tended to be part of more mixed and diverse networks.

One question arising from these findings is the feasibility of asking job-seekers in the coalfields to look further afield to increase their chances of finding work.  If people are to be encouraged to commute longer distances, for example to the nearby city, then more consideration needs to be given to the best way of achieving this, such as a measure of financial support to cover travel costs.  However, there are also many non-economic constraints that deter people who are only able to obtain low-paid work from commuting to cities.

Governance bridging coalfields and cities

A common feature of current approaches to sub-national economic development policy is the designation of particular centres or zones as the location for new economic activity, and the need for these to connect to areas of disadvantage.  Therefore, the study examined the nature and extent of collaborative governance arrangements and policy interventions that sought to provide a bridge of this kind between the coalfields and their neighbouring cities.

The evidence indicates similarities and contrasts between the three study areas.  Parallels exist in the local frameworks and instruments aimed at fostering economic development (involving increasing strategic co-ordination of district-level activities) and those geared towards community regeneration (involving intensive partnership-based working at neighbourhood level).  However, co-ordination between the two is generally limited in all three areas.  Wider collaborative structures are much further developed in South Yorkshire (including Sheffield) than in the Lothian-Edinburgh or Central Valleys-Cardiff areas.

While these South Yorkshire structures started to emerge with the growth of sub-regional efforts at co-ordinating economic development in the latter half of the 1990s, significantly the main catalyst was the designation, design and implementation of the Objective 1 Structural Funds programme covering the 2000-2006 period.  Since 2004 this has been enhanced by the use of sub-regional investment planning to allocate 80 per cent of Yorkshire Forward’s expenditure towards implementation of its Regional Economic Strategy.  The four South Yorkshire local authorities have had a strong input into this process.  Although inevitably guided by the concerns and priorities for their own areas, this active participation has largely been driven by the existence of a common planning and resource allocation framework that ties all parties together.  Currently this is being taken forward within the framework of the Sheffield City Region, which now also incorporates former coalfield areas in North Derbyshire and North Nottinghamshire as well, and appears to be following a similar polycentric approach to development planning.

The picture is very different in the other two areas studied.  In South Wales, there are embryonic collaborative structures emerging around a loose amalgam of local bodies under the banner of the ‘Capital Network’, now under the aegis of the Wales Spatial Plan (Welsh Assembly Government, 2006).  This covers the broader area of South-East Wales (Cardiff, Newport, and the Central and Eastern Valleys).  It is envisaged as “an innovative skilled area offering a high quality of life, international yet distinctively Welsh” (Welsh Assembly Government, 2004a: 49).  However, the role of the capital city, Cardiff, within these arrangements has yet to be fully specified, and it will take some time and effort to overcome the past years of mutual suspicion and antipathy between the city and the Valleys’ authorities.  Already key linking actions are being developed around skills enhancement of Valleys’ residents and widening of recruitment networks for Cardiff employers, as well as in other services such as waste management.  Greater collaboration in terms of economic development is emerging under the auspices of the South East Wales Economic Forum, and improving the speed and quality of transport services is being advocated by the South East Wales Transport Alliance.

At the same time, as a counterbalance to the strength of Cardiff and the coastal belt, a strategic framework for the Heads of the Valleys, Heads – We Win, was launched in 2005.  This covers Blaenau Gwent to the east, as well as the northern parts of the Rhondda, Rhymney and Taff valleys that are within the study area.  The aim is to develop “a strong, better balanced economy”, prioritising health, education and training, transport, the natural and built environments, and tourism (Welsh Assembly Government, 2005).  Interestingly, availability of funding for these proposed developments is being made contingent upon increased collaboration between the public bodies operating in the area.

Such activity was much less apparent in the Lothians-Edinburgh study area.  Here, it appeared that the immense strength of the capital city, Edinburgh, and its role as locus for major development projects promulgated by the devolved government based there mean that calls by the much smaller and weaker authorities in the Lothian coalfield for greater collaboration has tended to fall on deaf ears.  Although the Edinburgh and Lothians Local Economic Forum is charged with resolving overlap and duplication, improving partnership working and engaging the business community, its limited remit  means that its main role so far has been to improve the co-ordination of services for small firms.  Although its current strategy document (Edinburgh and Lothians Local Economic Forum, 2003) advocates a number of measures to build on links between Edinburgh and other parts of the city region, its lack of powers and resources mean that it has been unable to foster much progress along these lines.  In other words, the few cross-boundary structures that exist in Edinburgh and the Lothians tend to involve fairly loose partnerships that set strategic directions for other local agencies to follow, with little in the way of resources to help achieve their goals.  The bulk of public sector investment decisions still occur within a local authority framework, both in institutional and spatial terms.

This brief outline emphasises the fact that the strength of collaborative governance across the three areas does not match the pattern of economic change.  The analysis of labour market links presented above shows that the Lothian coalfield has the strongest economic interactions with its neighbouring city, Edinburgh, yet its sub-regional governance structures are weakest.  Conversely, in South Yorkshire the former coalfield has recently been outstripping Sheffield in terms of employment growth, yet South Yorkshire’s wider collaborative governance structures are the most firmly established.

It might have been expected that stronger collaborative governance at sub-regional level would have gone hand-in-hand with greater economic integration between coalfield and city – after all, promoting the city’s growth has for many been the rationale behind the city region approach.  In fact, stronger collaborative governance sits alongside weaker economic links in the three case study areas.

In the Lothian-Edinburgh case, the coalfield-city links are so strong as a result of normal market activity that extensive intervention by public sector agencies is  perhaps seen as unnecessary.  However, several interviewees observed that the lack of formal co-ordination makes it difficult to maximise the benefits of Edinburgh’s growth for the coalfield area.  In particular, it has meant that progress with attempts to connect employability schemes in the coalfield to job opportunities in the city has been slow.  Similarly, lack of co-ordination means that land allocated for employment in the coalfield is not being used to ease development pressures within the city boundaries.

In South Yorkshire, though, the competing requirements of different urban centres and employment zones call for stronger co-ordination.  There is an important political element to this, intended to deliver a spread of resources, developments and benefits across the sub-region, and the need to connect where people live to places of opportunity.  South Yorkshire comprises several urban centres, so a balance needs to be struck between its constituent parts – coalfield, city and other urban areas alike.  The crucial role of a major EU funding programme (Objective 1) in bringing all parties together in a way that sought to reconcile their different interests has already been highlighted.  Interestingly, recent developments here appear to endorse this, proposing more formal collaborative arrangements between existing partners as the governance pattern for the Sheffield City Region, rather than the ‘city region mayoral’ route recommended in the recent Local Government White Paper (CLG, 2007), and favoured by the Core Cities Group.

In South Wales, in contrast, there remains a stark contrast in economic fortunes between Cardiff and the coastal belt on the one hand, and the Central Valleys on the other.  This has made it difficult for local stakeholders to find common ground upon which collaborative governance could be built.  This has been compounded by the more disparate and fragmented pattern of governance that exists in the Central Valleys.  This echoes the findings of the recent review of local service delivery in Wales that noted the difficulties of delivering institutional change (Welsh Assembly Government, 2006).  As in Scotland, this relative weakness contrasts sharply with the political importance assigned to the capital, especially in the immediate aftermath of devolution in the late 1990s, and in particular the centralising tendencies of the WAG in its early years (Cooke and Clifton, 2005).  This has started to recede more recently, for example with the relocation of some sections of WAG to other towns like Merthyr Tydfil.  As the new national arrangements have begun to settle, then attention has increasingly been turned to breaking down the barriers that exist between local authorities and other agencies, with a view to securing a broader spread of economic benefits.  At the same time, the economic and physical geography of the Valleys is hardly advantageous in terms of securing vigorous economic regeneration.  Even if they had been in place, stronger city region governance structures alone could not be expected to overcome these formidable obstacles on their own.

Conclusions

Arguments in favour of city regions suggest that more co-ordinated structures of governance are associated with rising economic integration between the places involved.  However, in this study of three coalfield-city areas in the UK the converse appears to be the case.

Thus, the formal institutional links between the Lothian coalfield and Edinburgh city are the weakest, yet the labour market linkages are strongest and the area is functioning along the lines of a ‘monocentric’ city region.  In contrast, South Yorkshire has the best-developed network of collaborative and city region structures, but the weakest tendency towards city-coalfield economic integration.  This reflects the wider opportunities for coalfield residents provided by a number of employment centres across the sub-region.  South Wales has rising labour market integration between the Central Valleys and Cardiff, but from a relatively low base and against a backdrop of merely embryonic and slowly developing collaborative arrangements.  Here it will take some time to overcome the historic suspicion and long-standing rivalries between the coalfield and Cardiff, as well as between local authorities in the Central Valleys themselves.

This apparently inverse relationship between the extent of cross-boundary collaborative governance and the degree of economic interaction between the coalfields and their neighbouring cities may seem surprising at first glance.  However, further scrutiny reveals that there does appear to be a strong association between the two.  Thus, sub-regional co-ordination may be a less significant requirement in fairly compact, rapidly growing conurbations such as Edinburgh and the Lothians.  Where there is a larger polycentric economic geography and a relatively dispersed growth prospects, as in South Yorkshire, a greater degree of inter-agency collaboration may be required to maximise economic development opportunities and to ‘lock in’ potentially rival local agencies to a common policy-making and resource allocation framework.  In other words, the co-operation across the sub-region has been more about securing a balanced distribution of economic activity, rather than solely seeking to cement city-coalfield links.  The absence of similar structures for the Central Valleys and Cardiff is a product of the twin problems of an unbalanced economic geography and weaknesses in local governance.  The lack of progress with increased collaboration at sub-national levels in Scotland and Wales is also connected with the focus on the national scale in the immediate aftermath of the devolution settlement in 1997 (Cooke and Clifton, 2005; Goodwin et al., 2002).

An important implication of these findings is that the general economic trajectory of different coalfield areas depends primarily not on governance arrangements, but on other factors, including the market.  Clearly specific historical, geographical and political contexts are important conditioning factors in this differentiation.  However, this overall conclusion still has significant implications for current debates over reform of public-sector management structures.  The message is that these are perhaps less important than underlying economic dynamics and the availability of suitable policy instruments to exploit them.

In other words, the success of former industrial areas like the coalfields in adapting to economic change appears to depend mainly on issues like accessibility, location, inherited economic structure and general trends in economic activity.  The deployment of appropriate policy tools and sufficient resources to support them can serve to maximise the benefits that flow from these, in a way that inter-agency management arrangements alone could never do.

This analysis also suggests that the quest for rescaled territorial governance arrangements that can somehow ‘match’ the complex, fluid and diverse spatial networks of the modern economy, fragmented as it is into increasingly specialised segments and sectors, is at best misguided and at worst futile.  This is not to say that cross-boundary collaborative governance cannot improve existing conditions for securing new capital investment or major infrastructure developments.  Indeed, it would appear to have an essential role in mediating conflicts and debates in terms of public resource allocation and use.  This is particularly the case with respect to the broad distribution of economic activity across neighbouring places and jurisdictions.  However, its limitations and its inherently political nature should be clearly recognised.  At the same time, there should be renewed attention paid to the more appropriate local scales at which other forms of government activity (such as skills training and access to employment projects) work best.  Finally, there should be an explicit recognition that any attempt at co-ordinating the activities of several complex organisations across different geographical jurisdictions is inherently difficult.  Whatever governance structures are put in place, strategy formulation and decision-making will inevitably have a strong political element.  Whether such rescaled arrangements are supported by sufficiently robust political processes is a question that will require further investigation.