GaWC Research Bulletin 185

(A)

Clusters, Economic Geography and Regional Development Policy: Insights from the UK Broadcasting Industry¹

N. Pandit* and G. Cook**

 

INTRODUCTION

It is a commonplace that creative industries, of which broadcasting is an important example, are principally urban phenomena and have a strong tendency to be highly agglomerated within particular cities (Florida 2002, Hall 2000, Landry 2000, Scott 2000). This chapter will analyze the causes and effects of such agglomeration. Scott (2000) identifies New York, Los Angeles, London, Paris and Tokyo as the leading examples of privileged centres for the production of cultural artefacts , not least due to the interactions between a broad range of agglomerated cultural industries within them. Within television broadcasting, including programme production specifically, Hollywood is dominant with London and Tokyo being other major global centres .

The chapter focuses on television broadcasting which far more highly concentrated than radio broadcasting. In the course of the chapter there will be reflection on a number of key issues and debates in the economic geography literature in respect of which the broadcasting industry offers some relevant evidence. These debates are briefly set out in the literature review. The evidence reported thereafter relating to the UK broadcasting industry is based on three sources: a large scale econometric study of cluster dynamics using financial data on 1213 broadcasting firms throughout the UK; an interview survey consisting of 72 interviews conducted mainly with firms in the three cities, London, Bristol and Glasgow, between 2001 and 2002; a questionnaire survey of firms in London conducted between January and April 2004.

The UK broadcasting clusters may be compared with the large literature regarding Hollywood (Christopherson & Storper 1986, De Fillippi & Arthur 1997, Miller & Shamsie 1996, Scott 2000, 2002, Storper 1997, Storper & Christopherson 1987). Whilst this literature focuses on film production it is important to be aware that employment related to television in Hollywood is double that related to film and has been less subject to geographic dispersal (Verrier 2005). The evidence presented in this chapter indicates that the nature of clustering in broadcasting is akin to that identified in Hollywood and typical of that claimed to be characteristic of cultural industries more generally (Scott 2000).

LITERATURE REVIEW

What is the Cluster Really Like and How Much Can We Generalize about Clusters?

The literature abounds with different typologies of geographic concentrations of economic activity (McCann & Sheppard 2003). These typologies are important because they bring into sharp focus that not all clusters are alike. Moreover, the nature of the cluster has implications for its potential value in the context of a regional development strategy as well as indicating what types of policy action are likely to prove most of value. A number of important ideal types have been identified by Markusen (1996), each of which manifests a certain type of economic logic.

(i) The Marshallian (Marshall, 1920, 1927) New Industrial District (NID), of which the Third Italy is identified as an important variety, emphasised by many leading authors (Best, 1990, Piore & Sabel, 1984, Porter, 1990, Saxenian, 1994, Scott, 1988). The “Italianate” version differs from the original Marshallian conception in that there is argued to be a high degree of conscious co-operation among firms and also important support from local organisations and institutions. Similarly, Gordon & McCann (2000) distinguish between pure agglomerations which do not involve substantial interaction and social network type concentration which do.

(ii) The hub-and-spoke district where regional structure revolves around one or several major corporations in one or a few industries.

(iii) The satellite industrial platform comprised chiefly of branch plants of multinational corporations. Such districts are typified by a lack of interaction among the firms located there. Firms are above all dependent on corporate parents.

(iv) The state-anchored district where a major government tenant anchors the regional economy.

Is any of these descriptions adequate as a description of the broadcasting industry? There has been a degree of dispute regarding the nature of Hollywood as a cluster. The dominant view is that it is a flexibly specialised industrial district (Christopherson & Storper 1989, Scott, 2002, Storper 1997, Storper & Christopherson 1987). Against this Aksoy & Robins (1992) have claimed that Hollywood is (though they do not use this language) more akin to a hub and spoke with major studios calling the tune and independent production companies being highly dependent.

How Robust is the Clusters Concept as A Basic for Policy?

The idea of promoting clusters has become central to regional policy in the UK and more widely. Policy thinking has lighted upon two particular ideas about clusters. The first is the ideal type exemplified by Silicon Valley (Saxenian 1994) and the idea of flexible specialization exemplified by the Third Italy and Baden Wurttemberg (Best 1990, Piore & Sabel 1984, Scott 1988). The second is the cluster concept developed and popularized by Porter (1990) in which the idea of flexible specialization again features prominently. There have been two general criticisms of the flexible specialization model. Firstly, that it is not an accurate representation even of Silicon Valley, Baden Wurttemberg and the Third Italy (Malmberg & Maskell 2002). Secondly, that this type of concentration is not the most common and that other types exist which also have a distinct rationale (Gordon & McCann 2000, Markusen 1996). A particularly trenchant critique of Porter has been advanced by Martin & Sunley (2003). The nub of their critique is that the concept is too vague and elastic, failing to be specific about what the spatial limits to a cluster should be nor which sets of industries ought to be linked within a cluster. Martin & Sunley (2003) round their critique off with a trenchant discussion of the shortcomings of the cluster concept as a guide to regional development policy. All firms, whether clustered or not, may benefit from support and diversity rather than specialization may be more important to cluster development and regional adaptability.

The Connection between Localized Concentrations of Activity and Global and National Hierarchies of Cities

Several authorshave championed the region as the most important spatial scale over which clustering processes operate (Coe & Townsend 1998, Cooke & Morgan 1997, Scott 2001, Scott et al. 2001, Storper 1997). Amin and Thrift (1992) argue persuasively that the emphasis on local production complexes is overdone. Models which are locally based do not recognize the importance of emerging global corporate networks and interconnected global city regions (Scott et al., 2001). They argue that centres are needed within which to generate and disseminate discourses, collective beliefs, stories about what world production filieres are like and to develop, track and test innovations.

The relationship between the city and economic advantages of collocation is of particular relevance to the broadcasting industry which is almost everywhere an urban and specifically large city phenomenon. The relationship between cities at a global scale is also relevant given the presence of multinational media companies in a number of key cities. Finally the relationship between (generally) capital and lower level cities in the national city hierarchy is also relevant given the policy trend in several countries towards spreading broadcasting activity among the regions.

The “Death of Geography” and the Nature of Knowledge

The idea that improvements in transportation, communication and information technologies is leading to a frictionless world where distance is irrelevant and the need to physical collocate is of diminishing importance has been both widely aired (e.g. Cairncross 1997) and strongly criticized (Morgan 2004). Broadcasting appears a prime candidate for the “geography is dead” thesis because it produces a “weightless” product/service which can be delivered quickly, cheaply and reliably at any spatial scale. Morgan (2004) has recently produced a vigorous rebuttal based on a number of key arguments. Morgan’s first argument is that the “death of geography” perspective conflates spatial reach with social depth. Being able to receive data is not the same as being able to understand it. By extension, face-to-face communication carries highly important non-verbal expression which augments the spoken word. ICT can be useful for maintaining communities but much less for building them. Secondly, competitive advantage is based among other things on tacit knowledge, which being person-embodied and context dependent is spatially sticky. Gertler (2003) argues effective creation and sharing of tacit knowledge “…. depends on institutional proximity – that is, the shared norms, conventions, values, expectations and routines arising from commonly experienced frameworks of institutions.” (p91). Such institutions will be in part locally and in part nationally constituted.

Malmberg & Maskell (2002) argue that current theorizing on knowledge transfer does not explain why collocated firms are better at creating and transferring knowledge than either one large firm or a network of firms spread across a range of locations. This is matched by a dearth of empirical evidence illuminating how, in practice, knowledge is created and transferred. Malmberg & Maskell suggest that the vertical dimension of clusters has been overemphasized and that the horizontal dimension, the collocation of firms at the same stage in the value chain may be more important. A similar claim regarding media industries specifically is made by Bathelt (2005). They suggest that more avenues will be pursued by a set of independent firms than in one large firm and secondly that collocated firms tend to be well informed about what their rivals are doing.

SOME ECONOMICS OF TELEVISION BROADCASTING

Television broadcasting may be broken down into four stages: programme production; bundling programs into channels; physical distribution; final consumption and revenue collection. The predominant model for public service broadcasting, and to a somewhat lesser extent commercial broadcasting, has been a high degree of vertical integration. Television programme production has traditionally been broken down into three broad activities: pre-production which involves idea development, scriptwriting, casting, costume, location finding, set construction, financing and insurance; production; and post production (editing, laying of sound track, titling and special effects). Programme production is essentially a project-based technology in which these disparate sources of input and skills are pulled together for the life of the project. This project-based characteristic is shared with many industries, particularly in cultural and media industries (Grabher 2002). Two important characteristics are that these disparate activities, which often entail highly specialized skills and equipment, can be hard to coordinate under what are typically tight constraints of time and money (Jankowski & Fuchs 1995) and that the chain is only as strong as its weakest link (Caves 2000). A very high premium is placed on those who have the reputation of at minimum being reliable and at best able to make a show a sound prospect for success (Shew 1992). This, according to Caves (2000), places a premium on mechanisms which certify talent.

Demand for television programs in terms of viewers tuning in has the very important property that it is fairly stable and predictable, if slowly changing, in terms of total hours viewed, but very uncertain in terms of the popularity of new offerings (Dunnett 1990). As Scott (2000) argues, the organisation of production is becoming more geared to flexibility and innovation in order to cope with constantly changing and unpredictable consumer tastes. Collins (1988) suggests that the strongly ephemeral character of much broadcasting output means that the pace of innovation is much faster than in most consumer industries.

Economies of scale in programme production are not pronounced (Cave 1989). Economies of scale are significant in what may somewhat loosely be called “distribution”, which relates to the acquisition of broadcasting rights and bundling them into packages, typically in the form of a channel offering. Large distributors are able to absorb a large number of programmes which may be barely commercial and recoup on the relatively small numbers of hits making major media companies important agents organising financing, deal-making and distribution. The existence of large economies of scale and scope is of first rank importance. Firstly, the fact that broadcasters (and studios in the case of Hollywood) are large will create a natural physical agglomeration (Ellison and Glaeser 1997). Secondly, the large scale of broadcasters relative either to (most) independent production companies and final consumers brings elements of both monopsony and monopoly power, a feature given high prominence in Aksoy and Robins’ (1992) account of Hollywood.

GENERAL TRENDS AND INFLUENCES IN THE BROADCASTING INDUSTRY

An obvious point about broadcasting is that in every country it is subject to heavy government regulation which has generally had an influence on the location of broadcasting activity. The UK exemplifies general trends replicated the world over, not least since many national television broadcasting systems were set up to ape the British Broadcasting Corporation (BBC). Firstly, the state broadcaster was set up in the capital which remains an important strategic centre in the industry. Exceptions to this trend are countries which have political capitals other than in major economic centres. Secondly, when independent television was set up in the UK, a deliberate policy of ensuring that network production (i.e. the production of television programs that would be screened nationally) took place in the regions was pursued. Other countries have been more or less dirigiste on this point, but many have attempted it (Euromedia 1997). Some countries have from the outset encouraged substantial broadcasting and programme production activity at a regional level, the prime examples being Germany (Kleinsteuber 1997) and Austria (Trappel 1997), reflecting their federal structures, and Switzerland (Meier 1997) which likewise has a decentralized system of political governance as well as distinct linguistic regions.

Attempting to spread broadcasting activity around the regions once a highly concentrated industry has been established in the capital has proved difficult. The UK has imposed quotas for regional production which have to a substantial degree been circumvented by independent companies from London setting up branch offices outside the capital to win business under the quota, but still essentially orchestrating operations from London. At the same time many independent production companies in the regions have felt it necessary to set up offices in London in order to help secure commissions from the major broadcasters. This centripetal force of the capital is also exemplified in the case of Japan, where despite attempts to spread activity around the regions when independent broadcasting was set up, activity quickly gravitated toward Tokyo (Brown 1989, Ito et al., 1978, Kuratani & Endo 2003). This impulse toward concentration in a single centre has also been demonstrated by the way Los Angeles has progressively drained activity away from New York in the post war period (Storper & Christopherson 1987) as has Glasgow in respect of Edinburgh in the post deregulation era in the Scotland since 1990. Moreover, attempts to woo film and television productions away from Hollywood to other centres in the USA have largely resulted in failure (Scott 2002). The strong centripetal power of major broadcasting centres the world over is telling testimony against the “death of geography”.

A second state policy which has been important around the world and is again exemplified by Britain is privatization and deregulation. The major impetus for broadcasting deregulation in Britain came with the Peacock Report (Home Office, 1986). The report led directly to the Broadcasting Act 1990 which aimed at promoting consumer sovereignty (Collins 1990). Among the major provisions of the Act were (a) The BBC, the Independent Television (ITV) companies and the future operator of Channel 5 were all required to ensure that at least 25% of their total programme output was contracted to independent producers; and, (b) A system of competitive bidding was introduced for ITV franchises, with the franchise being awarded to the highest bidder subject to a quality threshold.

The requirement for competitive tendering led to radical downsizing among the ITV companies, some of whom nevertheless failed to retain their franchises. The creation of new ITV companies as a result of this process, which made few programs, acted as a further stimulus to independent production ( Renton 1994). The impact of the Peacock Report on the BBC was even more immediate (BBC 1992) and between 1986 and 1993 the Corporation shed approximately 7,000 staff (which represented just over a quarter of employees). Many of those made redundant from both the BBC and ITV companies established themselves as freelance operators or formed independent companies in the vicinity of these major broadcasters. Roughly coincident with these developments has been the rise of cable and satellite broadcasting, which now constitute roughly one third of broadcasting activity in the UK. This has to some extent undermined the market dominance of the BBC and the ITV companies and has created further demand for independently produced programmes. These two developments bear analogy with the Paramount Decision of 1948 which forced Hollywood studios to divest cinemas, thus enforcing vertical dis-integration, and the rise of television in the 1950s, which lead to the demise of the mass-production, highly vertically integrated Hollywood studio model in the post war period in the USA (Scott 2002, Storper & Christopherson 1987). Both were fundamental processes underpinning the development of a highly vertically and horizontally disintegrated and specialised agglomeration which once started continued under its own self-reinforcing dynamics.

The demand for programme content to fill satellite channels, and to a lesser extent cable, has led to a strong upsurge in demand for programmes and programme formats (Moran 1998), the overwhelming majority of which are bought from the United States (Doyle 1993, Dominick et al 1993, Hoskins et al. 1997). The US has a very dominant position in this trade for several reasons, two important ones being the relatively low cultural discount on US products and the fact that major US media companies have considerable resources and accumulated expertise in developing overseas markets both for films and television programmes.

EVIDENCE ON THE UK BROADCASTING INDUSTRY

Some Stylized Facts

Broadcasting is highly agglomerated, with an estimated 70% of employment in film and television concentrated in London, which also produces almost three-quarters of independent television programmes. Within this, there is a very strong concentration on an area of approximately one square mile in Soho. Programme production is dominated by the BBC which is responsible for around 36% of programme production. The independent television broadcasters account for around another 25%. Another 25-30% of the market is accounted for by about 50 medium-sized, mainly London-based independent production companies (many of which are owned by larger media concerns) and the remaining 10% is accounted for by around 500 very small companies. The dominance of London compared to other regional centres in the UK is indicated in tables 1 and 2 which show respectively absolute numbers of employees and location quotients in broadcasting and related media industries. On the one hand London is much smaller than Hollywood (which has an estimated 132,000 employed in television-related activity), but also more geographically concentrated. Central London includes Soho and West London is dominated by the headquarters of the BBC.

Recent econometric research conducted on the British broadcasting industry (Cook et al. 2001) has yielded results which indicate that the dynamics of industrial clustering are subject to positive externalities of collocation which are generated in a highly asymmetric way. Two types of model were estimated. The first, a growth model, estimated the extent to which cluster strength, measured in terms of employment in both the firm’s own line of activity (own employment) and in related lines of activity (other employment), either impeded or enhanced the growth rate of firms located within the cluster. In almost every case, cluster strength in the firm’s own line of activity enhanced the firm’s growth rate, whereas strength in related lines of activity diminished firm growth, suggesting a congestion effect. The second type of model was based on firm entry and investigated the extent to which cluster strength in sub-sectors within each industry either appeared to attract or repel entry of firms into each sub-sector. Entry into a given sub-sector is almost always deterred by existing cluster strength in that sub-sector and entry attraction typically emanates from other sub-sectors.

In terms of growth dynamics, the strongest positive effects of cluster strength in a firm’s own line of activity appear in programme production and the manufacture of broadcasting systems. These are followed by post production and programme distribution. The centrality of programme production and post production in cluster dynamics is reinforced by entry dynamics. Programme production is the only sub-sector in broadcasting where cluster strength within the sub-sector attracts entry of like firms. Programme production and post production also exert strong positive entry attraction on each other. Thus there is a powerful virtuous circle where cluster strength in each of these sub-sectors not only attracts entry but also leads to faster growth. Both of these sub-sectors attract entry from firms in other sub-sectors within the industry.

Table 1. Numbers Employed In the Audio Visual Industries By Sector and Region

Source: Skillset

Table 2. Location Quotients In the Audio Visual Industries By Sector and Selected Region

Source: Skillset

Social Capital and Knowledge Transfer

Figure 1 indicates the top 10 benefits of locating in London according to the questionnaire survey. Scores were based on a simple summation of the Lickert scale responses ranging from 0 (not applicable) to 5 (very important). This leads to a maximum potential score of 1020 (204x5). This figure immediately indicates the importance of social capital based on face-to-face interaction.

The ability to have face-to-face contact emerges as the single most important factor, closely followed by the ability to build and maintain personal relationships. Why should these two things be so important? Here the interview evidence across all three cities was strong and consistent. In large part it is due to the fact that what is being created is most importantly an image and sound which will convey meaning. These are cultural artifacts whose most important qualities are the extent to which they convey meaning, emotion and information. This makes it of paramount importance that those who are engaged in their production understand the meaning which is to be conveyed. In order to come to that understanding, communication with the maximum “bandwidth” is required, i.e. face-to-face contact (Pratt 2000, Scott 2000). A network of personal contacts allows teams to be assembled quickly in order to carry out what by most industry standards is a short-term project with the assurance that those engaged will be able to understand and produce what is required. Physical proximity in London promotes rich circuits of information about particular individuals which can be (and are) tapped into when assembling a team. These facts are salient reminders that the ability to produce is a vital influence on location decision, whereas much of the literature places emphasis on “ideas” and innovation as if they were disembodied from production. Moreover, as Pratt (2000) argues, “death of distance” arguments ignore the fact that many media products have a physical form, which in the case of broadcasting is highly significant as it must meet exacting technical as well as creative standards.

The advantage of physical proximity in promoting trust and cooperation with other companies was highly rated, 9 th out of 67 factors in the questionnaire survey. Trust looms particularly large in the relationship between production and post production and provides an important part of the reason why they are almost always co-located. Handing over the “rushes” (raw film or video which has been shot) for post production is to entrust another with a valuable asset in which you have invested heart and soul. To reduce this uncertainty, people like to work with those whom they can trust to “get it right”. The extent to which this potential advantage is seen to be important in practice is associated with location, rather than with line of business with the strongest association being location in Soho and to a lesser extent a wider definition of central London (important because advertising has a different centre of gravity to broadcast and film). This evidence is consistent with the transaction cost approach to explaining the value of proximity (Storper, 1997), subject to the qualification that what is feared is incompetence rather than opportunism, although Banks et al. (2000) do produce evidence that cultural entrepreneurs do get cheated and fear being cheated. The path of production is sometimes smoothed by personal favors. A common scenario which was often repeated during the course of the interviews was where, for example, an independent production company would be strapped for cash for post production work and would be able to arrange for a “special fee” from a post production company with the promise that on some future occasion the favor would be returned by paying a little over the odds. Another example would be engineers within the post production community lending each other equipment, usually on the implicit understanding that the favor would be returned should such a situation arise in the future. Such behavior has to be underpinned by trust, a system for establishing and maintaining reputations and norms of reciprocity.

Figure 1. Top 10 Benefits of a London Location.

The questionnaire evidence indicates knowledge spillovers are locationally specific within London. The importance of knowledge transfer is moderately high and again the degree of importance is differentiated among various types of firm. Post production companies and firms in Soho are significantly more likely to rate this factor as important. One interpretation is that denser concentrations of firms will experience larger amounts of knowledge spillovers because of greater frequency of interaction. Firms in Soho and in central London were significantly more likely to cite meetings in the street among their three most important types of informal interaction (over 7 times more likely to cite if in Soho and 4 times more likely to cite if in central London compared to firms outside those areas.) Firms in Soho and central London were also significantly more likely to cite meetings in restaurants, wine bars and pubs than firms outside these areas. On the other hand, firms outside central London were significantly more likely to cite interaction via telephone, e-mail and professional bodies. Thus the general pattern is that the closer firms are to Soho, the more likely it is they will find exchanges with colleagues at informal social settings or simply meeting by chance in the street to be beneficial. The significantly higher importance attached to collocation by firms in post production tallies with Coe’s (2001) observation that post production has been spatially sticky in Hollywood even at a time when production, at least for more standardised output, has been increasingly moved to Vancouver. Glasgow differs from Bristol in some important respects. The degree of cooperation and informal interaction among its independent companies is palpably lower, the lack of “untraded interdependencies” being remarked upon by Turok (2003). One possible reason for this is the much lesser degree of physical propinquity in Glasgow compared to Bristol. What these results again indicate is that firstly dense communication is important, but where and how that communication takes place is influenced in a fine-grained way by precisely where companies are located and what line of activity they are engaged in.

An important word used repeatedly in the London interviews was “community”. The same word was used somewhat less frequently in the Bristol interviews and far less frequently in Glasgow. There are several important facets of the existence of this community. Firstly, there are overlapping communities, a wider broadcasting community and a more specific community relating to a particular craft or profession to which an individual will belong. Secondly, access to the community is granted by certain rites of passage, typically a year or two doing menial work for little or no pay as a runner during which commitment, competence and attitude are assessed. This ensures appropriate socialization, imbuing the individual with attitudes, norms of behavior and linguistic competence as well as formation of particular craft skills. How to behave and interact with others is a highly important dimension of this inculcation and testing for suitability (Grabher 2002). Membership of the appropriate community is an important foundation of trust. The desire to establish reputation among peers is also an incentive to effort and innovativeness (Heydebrand & Miron 2002). Norms of information sharing and egalitarianism within communities are also important contributors to innovation.

The Labour Market

Labour market pooling is highly important. A pool of talented labour with relevant skills stands as the pre-eminent local factor helping firms innovate according to the questionnaire survey. This is the third highest ranked factor across the whole questionnaire. From the employer's point of view, London is an attractive location, despite the obvious congestion, because it offers such an exceptional pool of creative talent relative to other areas of Britain. There is clearly a dynamic at work where the reputation of particular regional centres, above all London, attracts talent which makes the centre a more desirable place to do business and so on (David & Rosenbloom 1990, Nachum and Keeble 1999). The strong pull of London on highly skilled labour is indicative of a much wider trend (Dorling & Thomas 2004, HM Treasury 2001). An important counter-example is Bristol, which is the premier location for natural history producers, based around the Natural History Unit of the BBC (Bassett et al., 2002). Here anyone who aspires to work at the highest level in this genre has to operate in this location.

Labour market flexibility was generally viewed as being a source of advantage in a variety of ways. The most important advantage was seen as the ability to recruit good people at short notice, which stood apart from the other advantages of a flexible labour market. This is clearly important in media businesses which has a predominantly project based mode of operation. Reputation mechanisms are essential in such labour markets as labour is highly differentiated in terms of both skills and ability, which is something well beyond labour pooling benefits based on lower risk of rationing on either side of the labour market. As DeFillippi and Arthur (1998) argue, social capital (knowing each other) and human capital (knowing your trade) are intertwined in the labour market and both matter in a project-based industry.

Vertical and Horizontal Linkages

The production of moving images and sound is a creative process and a premium will often be placed on the ability to understand and realize the idiosyncratic creative vision of the customer, whoever that may be, channel controller at a major broadcaster to account executive for a major corporate client in the advertising world to the micro-scale independent film or programme producer. Accordingly, all types of production companies are significantly more likely to rate the ability to find firms who will supply bespoke services as important. The broad mix of media industries concentrated in London supports the fine-grained division of labour which it has been argued is crucial dynamic of productivity growth and innovation. Highly specialised firms are viable in some cases because they can supply their goods/services to a range of different types of media company (Christopherson & Storper 1986).

Figure 2 gives an impression of which sectors are most closely inter-related. The figure was constructed by first classifying each firm to a sector on the basis of what it classed as its most important line of activity. The next step was to examine which sectors those firms stated was the most important with which it had inter-relationships. The arrowheads show the direction of the relationship with the arrowhead entering the sector rated as most important. The thickness of the lines gives a crude indication of the percentage of each sector’s number 1 rankings being accorded to each sector.

A number of features stand out. Firstly, the broadcasters appear to be at the hub of the cluster, which is only expected. It is also intuitively correct that broadcast production should value being close to broadcasters, who are their clients. In this respect it is important to note that other broadcast activities includes a number of freelance production managers and directors. Advertisers also emerge as being an important part of the cluster, unsurprising given that they are high budget clients of broadcasters and parts of the post production community. Post production itself emerges as being quite central to the cluster, providing essential services to all forms of production, some of whom are subsumed in the other media category (corporate video and commercials production). Other media appears to be central too, however this simply reflects the fact that a wide number of other activities have been rolled up into this catchall category.

Figure 2. Firm Ratings of Important Interdependencies by Line of Activity.

Companies do benchmark against each other in a way entirely consistent with the account of Porter (1990) and Malmberg & Maskell (2002). A close eye is kept on what commissions rivals win, who they engage on their production projects and the quality of the finished work, always with an eye to learning something of future competitive advantage. There is a conscious rivalry whereby firms and individuals strive to outdo their peers, raising standards and promoting innovation. Companies scrutinise each others’ financial information to evaluate whether or not they are earning better margins on their work with a view again to emulating their performance. Rivalry tends to be lesser among the smaller companies which were widely seen as offering each other mutual support. Rivalry is also tempered by allegiance to of individuals to their particular craft community. It is also evident that the plethora of small independent production companies is able to pursue a broader range of ideas than would occur within a single integrated broadcaster-producer (see Storper 1997 for similar observations regarding Hollywood). This wide experimentation is an important way in which major broadcasters cope with the vagaries of demand. The existence of a web of small, low-cost independent companies is also a means of lowering programme costs. In both respects the structure of the London broadcasting cluster mirrors the strategy adopted by the major studios in Hollywood for coping with the implications of more fragmented and unpredictable demand (Storper & Christopherson 1987)

As Bathelt (2005) suggests, hub firms play a number of important roles within a particular agglomeration. They establish basic ground rules for programme formats. They attract new businesses as they are important customers, not least since they favour local suppliers both of programmes and broader services. Traditionally, the BBC, and to a lesser extent the ITV companies have been important trainers and there is current concern that many high level skills are being lost as the BBC in particular has scaled back on this training role. Both the BBC and ITV companies have also since 1990 provided sometimes considerable assistance to fledgling independent companies. On the other hand, as Aksoy & Robins (1992) identify in respect of Hollywood, large broadcasters do have a considerable position of power relative to much of the independent production sector, in the majority of cases offering cost plus contracts for programmes which entail the independent company signing over the whole or the majority of the intellectual property rights to the broadcaster. This perpetuates the dominance of the larger media firms and comparatively few have broken through to become anything like large scale and independent producers.

London is home to many major overseas media multinationals and is important as a global node for a variety of reasons. It is a primary cultural metropolis which makes it attractive to creative individuals as well as being a melting pot of ideas (the two are interrelated). It can provide firms not only with ideas but also production capability sophisticated and flexible enough to meet any demand. The UK in particular and Europe are highly important markets for US exports and the UK is a useful beachhead for penetrating Europe. Glasgow has not been able to date to establish itself either as a major focus for demand or production for overseas broadcasters and media conglomerates. Bristol has only to a very limited extent relative to London, the exception being the natural history genre where Bristol is the leading global centre.

Innovation

It is important to appreciate that London is a vital global node in both technological and creative innovation and that these two innovation systems are intertwined. Both Soho production and post-production and the Thames Valley manufacturing clusters have the common distinction of being important nodes in a global industry. Indeed the entire broadcasting complex forms a neo-Marshallian node in Amin and Thrift's (1992) sense. For example, when new equipment is being developed it is typically beta-tested in a number of top post-production companies. Where a multinational company such as Sony is concerned, these test sites will be located in key centres around the world of which London is of prime importance. This on-site testing is important for two reasons, both salient in their ways for the innovation process. Firstly, very good feedback is obtained on how to develop the product ready for market. Secondly, the fact that a piece of equipment has been tested by a leading facility and endorsed by them is very important in driving sales for the product. Being a test site also has important advantages for the facility. They get first go with the latest equipment, before it is available to some of their rivals. Moreover, they have the opportunity to shape the development of the product closer to their own requirements. The ability to work on cutting-edge equipment also attracts top creative talent and helps retain them.

Social capital is highly important in the innovation system, and is exemplified by considering the role of engineers. Engineers form a highly important network. Particularly among engineers there are examples of knowledge trading which exemplify the processes analysed by von Hippel (1988). Engineers have two important professional affiliations. One is to their profession; the other is to the cluster. Both of these transcend narrow company loyalties and lead to co-operative behaviours which help sustain the cluster and drive its creative and innovative dynamism. Engineers, for example, within the post-production play a pivotal role. They understand the technical side, but also have great insight into the creative side of the business, therefore they can articulate what the customer is really after in a way it would be much more difficult to accomplish if engineers from the manufacturing companies were to attempt to talk to creative people direct. Engineers from the post-production houses are able to visit the R&D labs of the manufacturing companies due to physical propinquity which is beneficial to both sides. Occasionally this may lead to the post-production engineers going to help out the R&D team in the manufacturing firm. In this respect, occupational, relational and physical proximity are all important in the creation and sharing of tacit knowledge.

CONCLUSION

By way of summarizing the above evidence, a typology of British broadcasting clusters will now be presented. Broadcasting appears to be a blend of three of Markusen’s ideal types. Clearly the importance, almost since the inception of the industry in Britain, of the BBC, marks it as falling partly into the state-anchored model. The existence of dominant firms surrounded by fringe suppliers indicates elements of a hub-and-spoke type cluster. Firms supplying the BBC and the ITV companies have traditionally been in a very dependent relationship. The industry is still dominated by the major broadcasters even though many new types of broadcaster have entered the market and the major broadcasters themselves have been forced to vertically disintegrate. The strength of the broadcasters at the hub of the cluster is critical. They will determine the extent and stability of demand for programming which in turn drives the derived demand for labour. As has been argued above, the formation of the labour pool is a vital process in forming a dynamic regional concentration. Programme production appears to have many features reminiscent of a Third Italy variant of the NID. There is a high level of both vertical and horizontal disintegration, in the independent sector at least. As Storper & Christopherson (1987) have remarked regarding the post-war development of Hollywood, once the process of vertical and horizontal disintegration starts, it becomes a self-sustaining process as the cluster grows. In these respects broadcasting has very strong similarities with what is believed to be known about Hollywood.

National policy was primarily responsible for the formation of this industrial district. Firstly, as has been discussed, the Broadcasting Act (1990) ushered in a massive increase in the scale of independent programme production. Also of critical importance was the establishment of Channel 4 in 1982 as a broadcaster without its own in-house production capability. London was the natural home of this new broadcaster and it helped to develop the independent production and post production cluster which was able, therefore, to prosper more easily in the capital after 1990 than it did in the regions (Allen and Miller 1994, Lambert 1982).

Regarding the questions outlined at the start of this chapter, broadcasting in the UK has mapped in an imperfect way onto the typology of clusters identified in the literature review. In contrast to much of the literature, the questionnaire study revealed how important is the “microclimate” within the cluster, calling further into question how well any single model or typology can reflect the particularities of a given cluster. All seem to imply a greater homogeneity within a cluster than has been revealed to be the case here. The importance of particular cluster benefits such as informal knowledge flows and personal interaction vary significantly by precise location and line of activity. There are strong asymmetries regarding which companies generate positive externalities within the cluster.

Regarding the suitability of the cluster concept for policy, the evidence here supports the considerable skepticism voiced by Martin & Sunley (2003). Firstly, the idiosyncrasies of the cluster are too marked for broad brush nostrums to be useful. Secondly, classic business support has been effective in Glasgow and the absence of a “cluster policy” has been of little obvious detriment in London and Bristol. Thirdly, it is manifestly the case that local policy has been much less important in shaping the overall national dispersion of activity than national policy. Moreover, London has such powerful advantages given its status as a world city and as a key node in the global broadcasting industry that local policymakers have to be realistic regarding what can be achieved in second tier cities.

Finally, the evidence presented here is not only that geography is not dead, but that it is probably more powerful than imagined, centripetal forces being much in evidence. Firstly, knowledge transfer and information sharing as well as the formation of social capital have been shown to be powerfully centered on the highly dense heart of the London cluster in Soho and a rapid decay with distance is indicated. Trust, interpersonal relationships and face-to-face contact have been shown to be of abiding importance. Buzz matters for all the reasons articulated by Storper & Venables (2004) and its effect rises with geographical density. Secondly, labour market dynamics are centered on London and exert a very powerful growth dynamic. Propinquity does facilitate benchmarking as Malmberg & Maskell (2002) suggest, however, much of the important horizontal externalities take place across organizational boundaries within particular craft communities where trust looms larger than they imply. Contrary to Malmberg & Maskell’s claim, vertical linkages appear dense and fundamentally important to cluster strength.

REFERENCES

Aksoy, A. & Robins, K. (1992) Hollywood for the 21 st century: global competition for critical mass I image markets. Cambridge Journal of Economics, vol. 16, pp. 1-22.

Allen, R. and Miller, N. (eds.) (1994). Broadcasting Enters the Marketplace. London: John Libbey.

Amin, A. & Thrift, N. (1992), Neo-Marshallian Nodes in Global Networks. International Journal of Urban and Regional Research, V. 16: 571-587.

Banks, M. Lovatt, A., O’Connor, J. and Raffo, C. (2000). Risk and trust in the cultural industries Geoforum, vol. 31, pp. 453-464.

Bassett, K., Griffiths, R. and Smith, I. (2002), Cultural Industries, Cultural Clusters and the City: the Example of Natural History Film-Making in Bristol. Geoforum, vol. 33, pp. 165-177.

Bathelt, H. (2005). Cluster relations in the media industry: exploring the 'distanced neighbour’ paradox in Leipzig. Regional Studies, vol. 39, pp. 105-127.

BBC. (1992). Extending Choice. The BBC's Role in the New Broadcasting Age. London: BBC.

Best, M. (1990). The New Competition, Harvard University Press, Cambridge, MA.

Broadcasting Act (1990). Eliz 2, Ch. 42

Brown, D.R. (1989) Comparing Broadcast Systems. The Experience of Six Industrialised Nations. Ames, Iowa: Iowa State University Press.

Cairncross, F. (1997). The Death of Distance. London: Orion Business Books.

Cave, M. (1989) An introduction to television economics. In Hughes & Vines (eds.) Deregulation and the Future of Commercial Television op cit.

Caves, R.E. (2000) Creative Industries: Contracts between Art and Commerce. Cambridge, MA: Harvard University Press.

Christopherson, S. & Storper, M. (1989). The city as studio; the world as back lot: The impact of vertical disintegration on the location of the motion picture industry. Environment and Planning D: Society and Space, vol. 4, pp. 305-320.

Coe , N.M. (2001). A hybrid agglomeration? The development of a satellite-Marshallian Industrial District in Vancouver’s film industry. Urban Studies, vol. 38, pp. 1753-1775.

Coe , N.M. & Townsend, A.R. (1998). Debunking the myth of localised agglomerations: the development of a regionalised service economy in South-East England. Transactions of the Institute of British Geographers, vol. 23, pp. 385-404.

Collins, R. (1990). Television: Policy and Culture. London:Unwin Hyman.

Collins, R. (1988). The Economics of Television. London: Sage.

Cook, G.A.S., Pandit, N.R. & Swann, G.M.P. (2001). The dynamics of industrial clustering in British broadcasting. Information Economics and Policy, vol. 13, pp. 351-375.

Cooke, P. and Morgan, K. (1998). The Associational Economy. Firms, Regions and Innovation. Oxford University Press, Oxford.

David, P.A. and Rosenbloom, J.L. (1990). Marshallian factor market externalities and the dynamics of industrial localization. Journal of Urban Economics, vol. 28, pp. 349-370.

DeFillippi, R.J. & Arthur, M.B. (1998). Paradox in project-based enterprise: the case of film making California Management Review, vol. 40, pp. 125-139.

Dominick, J.R., Sherman, B.L. and Copeland, G.A. (1993). Broadcasting/Cable and Beyond. An Introduction to Modern Electric Media. 2 nd Edition. New York: McGraw-Hill Inc.

Dorling, D. and Thomas, B. (2004) People and Places. A 2001 Census Atlas of the UK. Bristol: The Policy Press.

Doyle, M. (1993). The Future of Television. Lincolnwood, Ill.: NTC.

Dunnett, P.J.S. (1990) The World Television Industry. An Economic Analysis. London: Routledge.

Ellison, G. and Glaeser, E.L. (1997) Geographic concentration in US manufacturing industries: a dartboard approach. Journal of Political Economy, vol. 105, pp. 889-927

Euromedia Research Group. (1997). The Media in Western Europe. Second Edition. London: Sage.

Florida , R. (2002). The Rise of the Creative Class. New York: Basic Books.

Gertler, M.S. (2003). Tacit knowledge and the economic geography of context, or The undefinable tacitness of being (there). Journal of Economic Geography, vol. 3, pp. 75-99.

Gordon, I.R. and McCann, P. (2000). Industrial clusters: complexes, agglomerations and/or social networks? Urban Studies, vol. 37, pp. 513-532.

Grabher, G. (2002) The project ecology of advertising: tasks, talents and teams. Regional Studies, vol. 36, pp. 245-262.

Hall, P. (2000) Creative cities and economic development, Urban Studies, vol. 37, pp. 639-649.

Heydebrand, W. & Miron, A. (2002). Constructing innovativeness in new-media start-up firms, Environment and Planning A, vol. 34, pp. 1951-1984.

HM Treasury (2001), Productivity in the UK: 3 – The Regional Dimension. London: HM Treasury.

Home Office. (1986). Report of the Committee on Financing the BBC, Cmnd. 9824 London: HMSO.

Hoskins, C., McFadyen, S. & Finn, A. (1997). Global Television and Film. An Introduction. Oxford: OUP.

Hughes, G. and Vines, D. (eds.) (1989). Deregulation and the Future of Commercial Television. Aberdeen: University of Aberdeen Press.

Ito, M., Shino, H., Kataoka, T. and Nagatake, S. (1978). Broadcasting in Japan.. London: Routledge & Kegan Paul.

Jankowski, G.F. and Fuchs, D.C. (1995). Television Today and Tomorrow. It Won’t Be What You Think. New York, New York: OUP.

Kleinsteuber, H.J. (1997). Germany. In Euromedia (1997) op cit.

Kuratani, M. & Endo, Y. (2003). Tokyo’s central role in the knowledge-based economy. Nomura Research Institute Paper No. 65.

Lambert, S. (1982). Channel Four. Television with A Difference? London: British Film Institute.

Landry, C. (2000). The Creative City. London: Earthscan.

Malmberg, A. and Maskell, P. (2002). The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering. Environment and Planning A, vol. 34, pp. 429-449.

Markusen, A. (1996). Sticky places in slippery space: A typology of industrial districts. Economic Geography, vol. 72, pp. 293-313.

Marshall, A. (1920). Principles of Economics, London:Macmillan.

Marshall, A. (1927). Industry and Trade, London: Macmillan.

Martin, R. and Sunley, P. (2003). Deconstructing clusters: chaotic concept or policy panacea? Journal of Economic Geography, vol. 3, pp. 5-35.

McCann, P. and Sheppard, S. (2003). The rise, fall and rise again of industrial location theory. Regional Studies, vol. 37, pp. 649-663.

Meier, W.A. (1997). Switzerland. In Euromedia (1997) op cit.

Miller, D. & Shamsie, J. (1996). The resource-based view of the firm in two environments: the Hollywood film studios from 1936 to 1965. Academy of Management Journal, vol. 39, pp. 519-543.

Moran, A. (1998). Copycat TV. Globalisation, Programme Formats and Cultural Identity. Luton: University of Luton Press.

Morgan, K. (2004). The exaggerated death of geography: learning, proximity and territorial innovation systems. Journal of Economic Geography, vol. 4, pp. 3-21.

Nachum, L. and Keeble, D. (1999). Neo-Marshallian nodes, global networks and firm competitiveness: the media cluster of central London. ESRC Centre for Business Research, University of Cambridge, Working Chapter no. 138.

Piore, M. and Sabel, C. (1984), The Second Industrial Divide: Possibilities for Prosperity, Basic Books, New York.

Porter, M. (1990). The Competitive Advantage of Nations. London: MacMillan.

Pratt, A.C. (2000). New media, the new economy and new spaces. Geoforum, vol. 31, pp. 425-436.

Renton , T., (1994). Broadcasting enters the marketplace: the keynote address, in: Allen, R. and N. Miller, (eds.), (1994), op cit.

Saxenian, A. (1994), Regional Advantage: Culture and Competition in Silicon Valley and Route 128, Harvard University Press, Cambridge, MA.

Scott, A.J. (2002). A new map of Hollywood: the production and distribution of American motion pictures. Regional Studies, vol. 36, pp. 957-975.

Scott, A.J. (ed.) (2001), Global City-Regions, Oxford University Press, Oxford.

Scott, A.J. (2000) The Cultural Economy of Cities. London: Sage.

Scott, A. (1988), New Industrial Spaces: Flexible Production, Organization and Regional Development in North America and Western Europe, Pion, London.

Scott, A.J., Agnew, J., Soja, E.W. & Storper, M. (2001), Global City-Regions in Scott, A.J. (ed.) Global City Regions, op cit.

Shew, W.B. (1992). Trends in the organization of programme production. In Congdon, T. et al. (eds.) Paying for Broadcasting. London: Routledge.

Storper, M. (1997). The Regional World: Territorial Development in A Global Economy. New York, Guildford Press.

Storper, M. & Christopherson, S. (1987). Flexible specialization and regional industrial agglomerations: the case of the U.S. motion picture industry. Annals of the Association of American Geographers, vol. 77, pp. 104-107.

Storper M., and Venables, A.J., (2004). Buzz: face-to-face contact and the urban economy. Journal of Economic Geography, vol. 4, pp. 351-370

Trappel, J. (1997). Austria. In Euromedia (1997) op cit.

Turok, I. (2003). Cities, clusters and creative industries: the case of film and television in Scotland. European Planning Studies, vol. 11, pp. 549-565.

Verrier, R. (2005). Movies, shmovies – T.V.’s taking over L.A. L.A. Times 19.8.05.

von Hippel, E. (1988). The Sources of Innovation. Oxford:OUP.

NOTES

^* Naresh R. Pandit, Senior Lecturer in Economics, Manchester Business School, Booth Street, West Manchester M15 6PB, United Kingdom, Tel. +44 (0)161 275 6492, naresh.pandit@mbs.ac.uk naresh.pandit@mbs.ac.uk

^** Gary A. S. Cook (Corresponding author), Senior Lecturer in Applied Economics, University of Liverpool Management School, Chatham Street, Liverpool L69 7ZH, United Kingdom, Tel. +44 (0)151 795 3708, g.cook@liv.ac.uk

¹ This research was partly funded by the Economic and Social Research Council, award number R000223258 and the British Academy, award number SG-36816. The authors would also like to acknowledge the contribution of Peter Swann and Skillset who provided employment census data.

Edited and posted on the web on 23^rd November 2005