The Towel and Bathrobe Cluster in Denizli
The secret of success lies in courage' (a businessman from Babadag)
Textile production starts with the main inputs - natural or man-made fibres, dyes and chemicals - and involves two major operations, namely yarn prep¬aration and fabric weaving. Although there are firms of all sizes operating at the various stages of production, larger firms have become increasingly dominant as the industry has become more and more capital-intensive. More than half of the output of the industry is used by clothing manufacturers. The rest goes to the manufacturers of household goods (for example bed¬clothes) and industrial goods (for example vehicle seat upholstery). Compared with the textile industry, the clothing industry is more labour-intensive and the technology used is less sophisticated. Distributors have become increasingly important in this sector due to the dominance of retail activities by large firms, a phenomenon that has affected the organization and geography of clothing manufacture (Dicken, 1998) since the Industrial Revolution. First Britain and then other developed countries such as Germany and France moved into the mass production of textiles, often in localized clusters. Later on the industry grew rapidly in developing countries and either stagnated or declined in developed ones. Despite this trend, however, several developed countries (including Italy, France, Germany and Britain) continue to be amongst the leading exporters of textiles and clothing.
In the case of towels and bathrobes, Turkey is amongst the leading exporters in this product group, along with China, Pakistan, Germany, Italy and India. During the period 1996-2000 Turkey was one of the top three exporters, accounting for around 10-20 per cent of total world exports of these products. The main destinations are the United States and the EU countries, especially Germany, Britain, France, the Netherlands, Italy and Belgium. Imports of towels and bathrobes into Turkey, on the other hand, have been negligible (ITC, 2002).
The leading 'textile cities' in Turkey have traditionally been Istanbul, Adana, Bursa and Izmir, but in the post-liberalization era they have been challenged by producers in the provinces of Thrace, Gaziantep and Denizli. The leading centres of towel and bathrobe production are the cities of Denizli and Bursa, with a higher concentration of employment in Denizli which accounts for more than 50 per cent of the country's total production of towels and bathrobes.
The Denizli cluster is highly export-oriented, and more than 60 per cent of its products are sold either directly or indirectly to international markets (Temel etal., 2002). The Denizli Chamber of Commerce values the city's textile/clothing exports at US$1 billion per year, or about 10 per cent of the total value of textile exports from Turkey (TB MM, 2001a). More than 60 per cent of these exports are towels and bathrobes (Akaydin and Ay, 1998). Denizli industrialists are very proud of their towels, which are used at Wimbledon and in five-star hotels throughout the world. Strikingly, over 40 per cent of Germany's imports of towels and bathrobes come from Turkey (ITC, 2002), most of them from Denizli. Terms such as 'the shining star of Turkey', 'the Anatolian tiger' and 'the Denizli miracle' are frequently used by researchers and policy makers to describe the city's impressive performance in international markets.
While much has been written about Denizli's success in this product area, the strategic management aspect of this has been largely ignored. This chapter will take a detailed look at the sources of competitive advantage of the Denizli cluster and the resilience it exhibited in the face of recent challenges, including the economic crises of the late 1990s and early 2000s and competition from other locations. The chapter will also discuss the underlying reasons for and the associated costs and benefits of clustering in this particular case. The analysis will enable us to identify key issues in the link between clustering and competitiveness, which will in turn make it possible to consider the competitiveness of clusters more generally.
Origins and historical developments
The beginnings of the cluster can be traced back to antiquity. Textiles that are estimated to be 2000 years old have been found in the region, ranking amongst the oldest found anywhere in the world. These early textiles were made of wool, suggesting the practice of animal husbandry. With the devel¬opment of agricultural activities in general and the cultivation of cotton in particular, there was a shift to cloth made of cotton (Karaalp and Batmaz, 1998). Denizli's location in the Menderes Valley, the historic gateway between the Aegean Sea and Eastern Anatolia, was very instrumental in the high level of development attained by this region in these early times (TC Denizli Valiligi, 1998). Historians claim that textiles took up the largest space in the historical bazaar of Denizli, and each of the subsectors of the industry (such as fabric, clothing and dyes) concentrated in certain parts of the bazaar (Gokce, 2000).
Denizli's brightest historical period was the Roman era, during which its already strong position in the production of textiles was enhanced. In fact 'trimita', a type of fabric produced in the region, was so famous that the city was called 'Trimita' by some (TC DenizIi Valiligi, 1998). There is also evidence that some of the material used for Caesar's clothing was imported from Denizli (Sozkesen, 1998). There was a relative decline in the region's economic pos¬ition during the Byzantine era, mainly because the centre of commercial activity shifted from the Aegean and Mediterranean to Constantinople (Istanbul) and its environs. Nevertheless textiles continued to be amongst the leading items produced in the region (TC Denizli Valiligi, 1998). It is also known that the existence of various types of plant and water in and around Hierapolis (Mutluer, 1995) laid the foundations for the development of a closely related industry: dye production (Gokce, 2000).
During the early Ottoman period the key location of the city, then known as Ladik, guaranteed a vibrant commercial life. The careful processing of high-quality cotton resulted in textiles of superior quality that were known for their durability as well as their beauty. Ottoman historians report that a certain type of cotton fabric produced in Denizli and decorated with golden yarn was unmatched anywhere in the world in terms of beauty and quality. Cotton fabrics produced in the region were preferred by the Ottoman aristo¬cracy and were popular in Istanbul's palaces (Mutluer, 1995). In fact Denizli fabrics and clothing were so unique and valuable that they were given as presents to Sultans and other high-ranking people. It is also known that fabrics produced in Denizli were exported to various Mediterranean countries during this period (Gokce, 2000).
In the fifteenth century, when new routes to the East were discovered, the region lost much of its importance. Moreover the capital of the Ottoman Empire shifted from Bursa to Edirne and then to Istanbul, which favoured the development of the regions of Marmara, Thrace and Western Black Sea. In addition the restrictions imposed on foreign merchants damaged international trade and caused a relative loss of position for some key ports in the Aegean, including Izmir, the negative consequences of which were transferred to nearby cities such as Denizli (TC Denizli Valiligi, 1998). Nevertheless, as it was located in a region that produced one of the highest-quality cottons in the world, Denizli continued to be widely known as a leading centre of textile production.
Like many other areas of the empire, Denizli underwent an economic downturn in the late Ottoman era (Karaalp and Batmaz, 1998). This is mainly attributed to the Industrial Revolution in the west, which, coupled with the extremely liberal trade policy followed by the empire at the time, seriously damaged the already underdeveloped indigenous industry. One direct effect of this on Denizli was that cotton yarn and fabrics began to be imported, the first large-scale imports of textiles taking place in the 1870s. As a result, the production of inputs such as silk, raw cotton, wool and dyes gained importance in the region's economy as they were in great demand abroad. Towards the end of the century, another noteworthy event for the Denizli economy was the construction of the the Izmir-Aydin railway, whose arrival in the region triggered a revival of economic and commercial activities.
With the outbreak of World War I which was followed by the Turkish War of Independence the few remains of Ottoman industry virtually disap¬peared. Although there was a slight revival after the wars, following the return of some master craftsmen, the extent of destruction, especially in terms of human life, had such an effect on Denizli that in the early years of the Turkish Republic economic progress was minimal (TC Denizli Valiligi, 1998). In 1927 there were 1581 enterprises in Denizli, 27 per cent of which were textile-related and 98 per cent of which had fewer than 10 workers (Mutluer, 1995). Family members, particularly females, participated in the production process while the male members concentrated on obtaining inputs and marketing the products, mostly via tradesmen (Erendil, 1998).
A noteworthy event in the mid 1930s was the emergence of cooperatives in Denizli, which laid the foundation for industrialization by enabling indi¬viduals and firms to join forces to acquire cotton yarn at favourable terms, and thus reduce their dependence on yarn merchants (Mutluer, 1995; Karaalp and Batmaz, 1998). There was quite an increase in the number of small textile cooperatives in the 1940s, mainly because a state enterprise, the Sumerbank yarn Factory, was given the task of supplying cotton yarn to manufacturers via the cooperatives (Mutluer, 1995).
In the 1960s the use of electricity became widespread, triggering the adop¬tion of electrical looms. The modernization impetus was reinforced when Denizli was included amongst the provinces that were given priority status in terms of development in the early 1970s, which meant an increase in public investment. The Denizli Dyeing and Printing Factory was founded in 1974 and provided high-quality cotton yarn treatment, dyeing and printing services to producers. Another novel development was that the remittances sent by the Turkish workers in Germany fed the private capital accumulation in Denizli that had been initiated by the cooperatives. The 1970s also witnessed a rapid increase in subcontracting relations amongst the cluster firms (Erendil, 1998).
An important upturn in the Denizli economy took place in the 1980s, when the textile industry engaged in a significant export drive, making good use of the liberal government policies of the period. Meanwhile importing the required machinery became much easier, further facilitating the transition from home/workshop production to larger, sometimes factory-based production (TC Denizli Valiligi, 1998). Following an initial learning period, a real increase in exports took place in the second half of the 1980s and throughout the 1990s. Bigger firms and sectoral foreign trade corporations were usually the first vehicles by which small and medium-sized enterprises in Denizli began to export. Denizli towels and bathrobes were in great demand in international markets, which prompted firms to increase their production of these product categories. Other items such as bedclothes and babies' clothing were also exported, but their share of total exports remained well below that of towels and bathrobes. An examination of the cluster firms' date of establishment highlights the importance of the post-1980 period in Denizli's development in that around 80 per cent of the enterprises were established from the mid-1980s onwards, more than half of these after 1990 (Temel etal., 2002). As Sengun (1998) argues, it was this process of interna¬tionalization that created the economically vibrant environment that exists in Denizli today. In summary, the revival of the ancient tradition of textile production began in the early 1970s, developed in the 1980s and expanded both nationally and internationally in the 1990s.
This section closes with a summary of the current general economic struc¬ture of Denizli in order to put the towel and bathrobe cluster into that context. According to a survey by the State Planning Organization, there are around 6500 industrial establishments in Denizli, more than 90 per cent of which are small firms with up to nine workers. The total number of workers is almost 50000, 65 per cent of whom work for the larger firms (ibid.). Table 5.1 provides a breakdown of employment in Denizli by economic activity, in which the dominance of textile, apparel and related businesses is clear. Other sectors with a considerable presence are tourism-related activities, metal-working and machinery production, cutlery, leather footwear, earth¬enware products, products of marble and glass, food and beverages (especially flour, non-alcoholic beverages, roasted chickpeas, pasta, olives, wine, spices, milk and animal feed) and furniture. It should also be noted that the variety of industrial activities has increased in recent years (Sarica, 1997). In terms of export success, the textile/apparel sector is followed by the metal-working and food and beverage sectors, in that order. The leading item with regard to imports is machinery, followed by a range of textile-related products. Imports are mainly from Germany, Italy, Japan and Switzerland.
Sources of international competitive advantage
Basic factors of production
The principal items in the total cost of a typical textile product produced by the cluster firms are raw materials (especially cotton), labour and energy. Around 30 per cent of the major inputs are imported from abroad and another 30 per cent from nearby provinces. The remaining 40 per cent, are procured outside the regional economy (Temel et al., 2002). Although Turkey is amongst the world's top ten producers of cotton and the best quality cotton in Turkey is grown in the Aegean region, where Denjzli is located, Denizli has to import around 30 per cent of its cotton yarn requirement as cotton production falls short of the domestic demand. However the local availability of high-quality cotton was of great advantage to the cluster in the early years of its development, and more recently it has enabled the firms to build up a reputation for quality.
A comparison of international labour costs in the textile industry reveals that the wage rate in Turkey, at about US$2 per hour, is much lower than in developed countries (for example about $17 in Italy and $22 in Germany) but higher than in some other developing countries. For instance in China, India, Pakistan and Indonesia the rate can be as low as $0.5-0.6 per hour (SPO, 2001). In Denizli it is not uncommon to employ uninsured home workers, who are willing to work long hours and demand relatively low wages, providing cost advantages for the smaller firms (Eraydin, 2002a). However the fact that there is a high demand for workers increases their bargaining power. One of our interviewees, a manager of one of the leading cluster firms, said that his boss had warned him about this issue on his first day in office: ("Treat the workers well", said my boss, when I first started to work here. The unemployment rate is very low in Denizli; if they quit, it won't take much time for them to find a new job.' The shortage of qualified personnel is another problem, specifically managers, engineers, designers and marketing specialists. The recent establishment of apprenticeship centres and an occupational high school in the organized industrial zone are positive developments in this regard. Our interviewees emphasized the importance of improving the quality of life in DenizIi and promoting the positive aspects of the city in order to attract suitable personnel. 'If we manage to attract qualified human resources', one interviewee argued, 'the life span of the cluster will be longer, which in turn will help to realize our dream: to make Denizli the capital of textiles.'
The use of earnings from previous commercial activities and family assets to finance the founding of an enterprise is very common. In a survey conducted by the State Planning Organization, more than 90 per cent of the enterprises in Denizli reported that the owners' equity had been used to finance the initial investment, fal1ing to around 80 per cent for subsequent investments (Temel etal., 2002). The main reason for the low utilization of bank credit was its high cost. The fact that the use of short-¬term credit created severe problems in subsequent years when the Turkish economy was hit by a series of crises further increased the perceived risk of using bank credit (TBMM, 2001a). Financing is therefore a major problem for the cluster.
In general the infrastructure in Denizli is poor with the exception of the industrial zones. In particular, manufacturers complain about the poor quality of the motorways that link Denizli to other provinces, the high cost of energy (the highest among OECD countries) and the severe shortages of energy (ibid.) The airport is far from the city centre and the number of flights is limited. Although there is a railway to the Izmir port, there is no station close to the industrial zone. In short, infrastructure-related problems pose a threat to the long-term development of the duster.
With regard to factors of production, then, Denizli enjoys advantages in terms of raw materials and labour costs vis-a-vis developed countries but not compared with some other developing countries. The principal disadvantages are the lack of qualified human resources, the cost and shortage of energy, the cost of capital and the poor infrastructure.
Pressures to upgrade: firms' strategies
As noted earlier, more than 90 per cent of the cluster firms are small enterprises with up to nine workers (Temel etal., 2002). These firms usually work as subcontractors for larger firms that are engaged in international trade, and increasingly for multinational enterprises (Akaydin and Ay, 1998). As a result, large firms with 200 or more employees account for more than 60 per cent of Denizli's total textile exports (Erendil, 1998, pp. 209-10). The firms employing ten or more workers account for more than 90 per cent of value added (Temel et al.,2O02).
Family firms dominate the cluster. According to the manager of a leading towel firm who was preparing to transfer the management of the firm to the next generation of family members, there are many advantages to being a family firm: 'There is a clear consensus on what we should focus on: quality and customer satisfaction. It makes us faster and dynamic. I am the owner-¬manager of the company and have 30 years' experience in this sector. The firm and I are now ready to deliver the accumulated experience and created values to the second generation, to my sons.' Another manager pointed out that family members are dedicated and emotionally tied to the business, which increases their motivation and productivity: 'It is hard for professional managers to keep up with our pace of doing business here. Our tempo is such that we often work until very late, and sometimes for 24 hours.' In a similar vein, another interviewee underlined the high demands made on professionals: 'We all run here and expect professionals to run with us too.' Others, however, were unsure about the advisability of family dominance of the management process. According to one interviewee, for example, family firms are managed rather unprofessionally, and as the firm grows this lack of professionalism creates significant managerial problems.
The dominance of the family often means that the management structure is rather hierarchical and necessitates obedience to the older members of the family in general and the boss in particular. For the younger generation and non-family members, it is difficult to challenge the decisions taken by the older generation, who are also reluctant to delegate authority. Under these circumstances, ill-advised policies can easily become the norm. Although some firms have recognized the importance of good management, particularly in respect of marketing and strategy, and have taken steps to 'professionalize the farnily,’ by encouraging second.-generation members to take university courses in related subjects, this is still an exception. One reason for the slow pace of professionalization in general is simply that it is hard to find qualified professionals. Also, owner-managers might be reluctant to trust the judge¬ment of professionals in respect of key issues that might affect the future of the enterprise, such as new investments.
Partnerships are typically formed by members of the extended family, but such partnerships run the risk of breaking down when the members marry and start another family. Also, as family businesses grow in size and income, conflicts can emerge amongst family members, resulting in spin-off firms. Partnerships among friends and fellow townsmen are also common. The general tendency, however, is for such partners to establish their own busi¬nesses as soon as they feel financially confident and have built up a network of business contacts (Erendil, 1998). Spin-off firms can be in the same product area as the original firm or in a related area, depending on whether the sepa¬ration has been friendly or hostile. A similar dynamic may pertain when a professional leaves the firm to establish his or her own business. A final form of new business formation is when a financially successful firm in an unrelated sector decides to expand into the textile business. All spin-off firms and new entrants are perceived as posing a threat to the existing firms in the cluster, since new firms tend to steal both workers and customers.
The entrepreneurial abilities of the owner-managers of the family firms are one of the principal distinguishing features of the cluster. Indeed Denizli has a cultural heritage of supporting entrepreneurship, and the entrepre¬neurs of Denizli are proud that the city's industrial development has been achieved without notable government support. However it has been argued that envy of the heavy public investment in the nearby provinces of Aydin and Isparta motivated the Denizli firms to develop their business (Eroglu, 1998). In particular the establishment of a textile factory in the nearby town of Nazilli, in line with the etatist policies followed during the 1930s, caused annoyance in Denizll as it was thought that Denizli's long experience in the industry made it a better candidate for the factory. In retrospect, however, the managers interviewed were of the opinion that it had proved beneficial for Denizli as many of today's entrepreneurs might have gone to work at the factory and never realized their potential (Sengun, 1998).
An illustrative example of the numerous entrepreneurial initiatives taken by Denizli businessmen was the establishment of the city's second industrial zone. After a proper analysis of the situation, some two hundred entrepreneurs decided that it would take too long to complete the project if they relied solely on the government. In fact the average time taken by the government to finish such projects was nine years. Therefore they decided to use their own means to build the zone, the first private venture of its kind in Turkey. The zone was completed within a year. It is noteworthy that, despite the lack of government involvement, great effort was still required to overcome bureaucratic obstacles.
There are many other examples of the culture of self-sufficiency in Denizli. During the War of Independence, for instance, Denizli formed its own army and resisted occupation. Also, the first joint stock corporation in Turkey was established in Denizli. Moreover the people of Denizli built their own hospital, several high schools and even the governor's residence. Finally, in 2002 the firms in the industrial zone established their own power generation plant to overcome the energy-related problems they faced.
Denizli businessmen are reputed to have a strong work ethic, are highly committed to their work, associate hard work with success and enjoy working hard (Uslu, 2000). A common joke in Denizli is that when three people from DenizJi get together they talk about establishing a new business, whereas people from the neighbouring province of Aydin talk only about what to drink. Especially for the first generation of Denizli entrepreneurs, saving and not wasting a penny are important principles (Eroglu, 1998). According to one interviewee this had a negative impact on product quality for a while as owner-managers tended to postpone new investments. The following adjectives were repeatedly used by our interviewees to describe Denizli's entrepreneurs: dedicated, hard-working, motivated, confident, practical, sharp, ambitious, honest, patient, bold and determined, but also envious and merciless. Accordingly they have an optimistic outlook, are willing to take risks, can 'smell' opportunities and are quick to find solutions to problems. One can also argue that personal envy, as defined by Schoeck (1966) and Klein (1975), might have a triggering effect on motivation and innovation in this cluster. More than 80 per cent of the cluster firms now use state-of-the-art tech¬nology (Temel et al., 2002). A recent sign of the importance attributed to technology was the setting up of a technopark project (Tekmer) in the Pamukkale University campus. Another initiative, the first of its kind in Turkey, is the on-line exchange of second-hand textile machinery, set up by the Denizli Chamber of Industry. These machines are usually sold to other Turkish regions with some presence in textile-related businesses. Although the machines are still in good working order, Denizli industrialists target the upper end of the market, which requires the best and latest technology, making the frequent replacement of machinery inevitable. Although the current level of R&D activities is rather low - less than 5 per cent of Denizli firms conduct R&D on a regular basis (ibid.) - there is growing interest in it, usually in relation to product quality and control, product variety, new production technologies from other firms (domestic and foreign) and new product design. An interesting example was provided by one of the interviewees, who stated that cluster firms had worked hard to develop a towel that was both soft and absorbent, and had invented a machine that fulfilled this need. Relevant publications and catalogues, domestic and international fairs, and trade and governmental institutions are the main means of keeping up with developments and innovations in the sector. The final and most striking innovation-related observation for the purposes of this study is that around 60 per cent of the cluster firms closely follow other firms in the cluster when trying to learn about and keep up with innova¬tions in the sector (ibid.). (We shall return to this subject when discussing the reasons for geographic concentration.)
The cluster firms keep a constant eye on quality standards (ibid.), and managers confidently claim that they are able to 'produce high-quality products at competitive prices and sell them to difficult customers all around the world' (TBMM, 2001a). The importance placed on quality means that firms prefer to maintain long-term relations with trusted subcontractors. The share of subcontracting firms in the cluster is about 70 per cent (Eraydin, 2002a, p. 146). Subcontracting provides flexibility and enables firms to accept orders beyond their capacity. It is mostly the later stages of the production process (such as clothing production. ironing and packaging) that are sub¬contracted. For small firms, almost 50 per cent of production is associated with subcontracting; the corresponding figure for large firms is about one fifth (Temel etal., 2002). There are also less formal arrangements among cluster firms; for instance it is not rare for accessories to be bought or borrowed from competitors if a firm should urgently need them. One of the firms in our study once lent workers for a couple of days to another firm with which it had close relations. On a later occasion it obtained the help of that firm in having its products ironed when its own ironing facilities broke down. Such informal, ad hoc cooperation is common among relatives, friends and fellow townsmen.
Organized institutions for cooperation include the Aegean Garment Producers' Association (EGS), which was established in 1993 by 464 busi¬nessmen, 60 per cent of whom were from Denizli (Eraydin, 2002a). The EGS developed rapidly and helped its member firms in the areas of procuring inputs, exporting the final products, training, financing, insurance and organizing fairs (Uysal, 1998). lt has been especially successful at promoting members' exports; in fact the EGS Foreign Trade Company became the leading export firm in Turkey in the late 1990s (ibid.). The previously mentioned examples of the construction of the second industrial zone and the estab¬lishment of Birlik Enerji should be repeated here to underline the extent of cluster firms' cooperation if they are convinced that a project will benefit them all. Business associations are another example of organized mechanisms for cooperation, including sectoral organizations, the Denizli Chambers of Commerce and Industry, and industrialists' and businessmen's associations.
The presence of so many informal and organized mechanisms for cooper¬ation does not mean that competition amongst the cluster firms is not intense. In fact it is the rule for firms operating in the same area of business (Eroglu, 1998; Ilgeri, 1998). Our study found that cluster firms tend to conceal information on their products, markets and the companies with which they have relations. This applies even to close friends and family members who have separate establishments, and in fact is not surprising given that imitation is rapid in clusters. Nevertheless it is difficult to keep secrets, given the high rate of circulation of workers and other personnel, the use of common suppliers and the sharing of customers (Erendil, 1998). Another aspect of the issue is that if information on products, technologies and markets is a crucial asset for the cluster firms, it follows that employees and/or managers who are equipped with such information are indispensable and indeed firms com¬pete fiercely for them. The local environment is therefore characterized by competition that is balanced by cooperation and relationships based on mutual trust, the latter being especially important among firms operating in complementary areas (Eraydin, 2002a).
Adopting a copycat strategy to try to match the success achieved by similar firms in the cluster is very common and reflects the general attitude in the Turkish business environment. As Eraydin (2002b) notes: 'Turkish businessmen believe what they see.' Nevertheless there is some recognition in the cluster of the difficulty of sustaining such a strategy, since successful imitation could push firms into price competition (TBMM, 2001a). Relatedly, it has been argued by Sozkesen (1998) that Denizli may soon lose its competitive advantage in towel and bathrobe production since costs are likely to increase. What is implicit in this view is, of course, an evaluation of the cluster firms' competitive advantages which is argued to rely largely on cost advantages such as cheap labour. However, the cluster firms have perceived product quality (77 per cent) and labour cost advantages (56 per cent) as their key strengths, followed by production capacity (50 per cent). Their main disadvantages are government-related issues such as energy prices and regulations, plus financing and transportation costs (Temel etal., 2002, p. 101).
In summary, with regard to firm strategy the conditions are largely favourable in Denizli, but in order to maximize their potential and compet¬itive advantages the cluster firms need a better resource base, especially in respect of human resources.
Related and supporting industries
Sectors related to the towel and bathrobe industry in Denizli are shown in Figure 5.1, together with associated institutions. Table 5.2 provides data on exports of towels, bathrobes and other textile products that are competitive in international markets.
Paralleling the development of the towel and bathrobe sector, a demand for high-quality inputs, dyes, machinery, packaging and other services emerged in Denizli. Between 1965 and the late 1990s the number of factor¬ies producing dyes for the industry, for example, rose from zero to 20 (TC Denizli Valiligi, 1998). Local producers of textile machinery have achieved some successes (Eraydin, 2002a), a good example of which is a towel-making machine invented by one of the leading firms in the cluster. However these two sectors are at the development stage and are not competitive in international markets. Their full development will probably take some time¬ as one interviewee nicely put it: 'It is not that easy to be the master of a business; there must be a history of it.' Another interviewee argued that the best way for Turkey to develop its presence in these sectors would be via multinational enterprises since it was too late to start from the begin¬ning. Currently, 80 per cent of the machinery used by the cluster firms is imported (Temel et al., 2002). Some improvement is also needed in logis¬tics, transportation and banking services. The latter will be discussed in the following section as it is directly related to the macroeconomic conditions prevailing in the country, and thus to the role of government.
Amongst the non-textile industries that have a presence in Denizli, the tourism sector has played a key role in the success of the towel and bathrobe cluster. Pamukkale, which has been designated by UNESCO as a world heritage site, is a major attraction for the more than 1.5 million tourists who visit Denizli every year. The tourism sector has a multiplier effect on many other sectors, and it was argued by the interviewees that the large rise in tourism in the 1980s contributed positively to the development of the towel and bathrobe cluster in general and the process of internationalization in particular.
Overall, it can be argued that most of the industries that are related to and support the towel and bathrobe industry are internationally competitive in the textile city of Denizli. The exceptions are the dye and machinery sectors, which are still developing. Support from institutions such as industry¬ specific research and education establishments is available although there is room for improvement, especially in respect of the quality of the services they provide. Finally, the contribution of the tourism sector is noteworthy in that it has facilitated the internationalization of demand.
The role of government
As mentioned earlier, given the macroeconomic situation in Turkey, firms prefer - or rather are forced - to use their own resources for financing, which is limiting their growth prospects. Therefore one of the most pressing problems experienced by the cluster, financing, is in many ways related to the part played by the government. The only positive role assumed by the government in this regard is the provision of Eximbank credits. Although they are limited in amount and there are some concerns about the allocation procedure, Eximbank credits have benefited the cluster. However it is difficult to identify any other way in which the government has helped to ease the financial problems of the cluster firms, although there is no shortage of examples in the opposite direction. In February 2001, businessmen felt that the government had deceived and misguided them since the fixed exchange rate policy, which was supposed to continue for some time, was abolished that month, causing the worst financial turmoil in the history of the Turkish Republic. Many businesses paid a heavy price, and some had to close down their operations (TBMM, 2001a). Even worse, in the aftermath of the crisis there was a loss of mutual trust amongst the government, banks and indus¬trialists, and this is proving difficult to rebuild. Moreover the fact that the actions of the government have been rather unpredictable has made it a real challenge for firms to conduct any serious strategic planning. It was frequently stressed by the interviewees that it is essential to achieve political and economic stability, which is perceived as one of the necessary conditions for sustained growth.
There have, however, been a few positive government measures. Incentives provided during the 1970s, for example, triggered the development of the cluster, and in later years the establishment of the first industrial zone in Denizli contributed to the cluster's growth. More importantly, the export-oriented development strategy followed by the government since the 1980s has been especially beneficial for the cluster (Kucuker, 1998) as firms have been able to make good use of the financial incentives and tax rebates provided in accordance with the related policies. The interviewees agreed that the incen¬tive system introduced in the 1980s had worked perfectly in the case of Denizli. Finally, the establishment of the Denser free trade zone in 2000 served to boost exports.
All this, however, does not change the fact that in the absence of a strategic master plan, the help provided by the government is sporadic and policies are rather arbitrary. Moreover bureaucratic obstacles cause setbacks. Of par¬ticular concern is the time taken by Ankara to respond to requests. As one interviewee put it, 'the bureaucracy continues to crawl, while the industrialists are running'. Managers also complain about the improper implementation of regulations, corruption in the customs department and the unfair com¬petition that results from the existence of the grey economy. The factor-related problems mentioned earlier, such as the high cost of financing and energy, are other problems that are largely seen as government-induced.
Denizli businessmen are rather proud of their heritage of independence and expect little more from the government than the removal of obstacles: 'All we need are inputs and energy at competitive prices and proper financing mechanisms. This will enable us to compete on the same terms as our international rivals.' 'What we ask from the government is proper governance and economic stability. The rest we can handle ourselves.' Indeed DenizIi entrepreneurs typically establish their own businesses, gen¬erate employment, engage heavily in exports and pay their taxes (Goksu, 1998). As mentioned above, for example, they organized and completed the second industrial zone (known as the Ozdemir Sabanci Organized Industrial Zone) within a year, whereas it had taken the government 13 years (from 1976 to 1989) to finish the first one (ibid.; Sanca, 1997). In fact the phrases, 'growing without government help' and 'doing what the government would not do' are frequently heard in Denizli. However this does not mean that government cooperation is unimportant. As one manager warned, 'If the government continues to disregard our problems, it might be unrealistic to expect Denizli to raise new generations of entrepreneurs' (TBMM, 2001a). Having recognized this, the Denizli Chamber of Industry set up an office in Ankara to hasten bureaucratic procedures, engage in lobbying activities and foster better communication with members of government circles.
Reasons for geographic concentration
The landscape of the textile/apparel industry in Denizli
The towns of Babadag, Buldan and Tavas have played a key role in the development of the Denizli cluster (Figure 5.2; see also Figure 3.3 for the location of the province). Babadag in particular is a historically important textile production centre. In the 1940s, textile producers started to migrate to Denizli from these towns to seek better lives, bringing their skills and crafts with them (Sengun, 1998). The first corners were from Babadag and were known for their industriousness and dedication to business.
Located on the slopes of a mountain, Babadag had no arable land so the people of the town had to engage in non-agricultural activities. Given the age-old history of textile production in the region, the people focused on this sector and became either merchants or producers. During the early years of the republic, when there were no large textile factories in Turkey, Babadag assumed the important task of satisfying the domestic demand for a variety of textile items. A noteworthy event in Babadag during this period was the very first attempt in Turkey at quality assurance and standardization. Long before the national standards institution (TSE) was established, the Babadag Chamber of Commerce made it compulsory for its members to have the quality of their products approved by the chamber before they were sent to market. There was also a custom of master craftsmen signing their names on their products, which can be seen as an early version of branding. The items produced and traded were domestic products such as bedclothes and tablecloths.
Electricity was brought to Babadag in 1932 by an entrepreneur, Mehmet Islak, who installed a small power generation plant that was later enlarged by another individual, Necip Mutcali. This private initiative provided elec¬tricity to the populace only at night; during the daytime the limited supply was used to power the new electrical looms. Despite this extraordinary effort there was too little electricity to meet the needs of the textile producers, and in the 1940s this, together with the fact that the mountain terrain allowed little space for those who wanted to expand their business, resulted in the leading textile producers and merchants of Babadag migrating to the central town of Denizli in search of better opportunities. Kucuker, the founder of one of the leading firms in the Denizli cluster, recounted that in 1948 his family was one of the very first families to emigrate from Babadag to Denizli, and that the main reason for this was the need for a continual supply of electricity. Although Babadag eventually secured a proper supply in 1964, this time provided by the government, Babadagians continued to migrate to Denizli to expand their operations.
In the 1970s the first attempts to internationalize were again driven by entrepreneurs. One interviewee, an entrepreneur from Babadag, reported that in the 1970s he and a few friends went to Paris by car to seek business contacts there. When the government introduced liberal policies and export incentives in the 1980s the entrepreneurs were, as one interviewee put it, 'well prepared' and 'full of can-do spirit', so they made very good use of the incentives offered. The extent of their success and the contribution they have made to the development of the Denizli cluster is such that the label 'born in Babadag' guarantees a very good reputation in the cluster. One Denizli-born interviewee underlined the importance of this by stating that those managers who had not been born in Babadag (including himself) joked 'about applying to change their birth certificates so that they could reap the associated benefits.
s jn the case of Babadag, textile production has deep historical roots in Buldan. Historians report that textile items produced in Buldan during the Ottoman era were of high quality and demanded by the aristocracy. The cooperatives established in the 1940s played a key role in the development of the textile sector in Buldan by facilitating input supply and product marketing. Production increased further when second-hand electrical looms were purchased from Bursa and Sumerbank factories in the subsequent years. Currently there are 1112 electrical and 135 manual looms in the town, and the sound of weaving emanates from almost every house, just as in Babadag. Buldanians continue to produce hand-made textiles of unique quality, and use natural dyes prepared in accordance with secret recipes passed down the generations.
Another textile town in Denizli province is Tavas, which is home to numerous tailors specializing in clothing for men and children. Although textile production has well established roots in Tavas, the presence of a disproportionally high number of tailors is associated with what can best be described as a historical accident (Krugman, 1991a). The origins of this phenomenon can be traced to a Greek tailor who moved to Tavas during World War II and taught the details of the craft to Tavasians, who were already skilled at textile-related activities. Nowadays the tailors mostly target the price-sensitive segment of the Turkish market. Textile workers in the Kizilcaboluk district of Tavas specialize in the production of various items for the home, plus handkerchiefs and hand-made garments. The master craftsmen of Kizilcaboluk excel in arty designs and work as subcontractors for larger Turkish firms that target the upper end of the clothing market (such as Beymen and Vakko).
The towel and bathrobe cluster is predominantly concentrated in and around the provincial capital of Denizli. The city hosts merchants and numerous small clothing firms operating in workshops, business centres and even apartment buildings. Some of the larger enterprises are located in the Sumer district in the north of the city, others are situated next to the motorways connecting Denizli to Izmir and Ankara (Mutluer, 1995). In the late 1980s some of the larger firms moved to the industrial zone on the out¬skirts of the city. Now, 90 out of the 134 firms operating in the zone are in textile-related businesses (DTO, 2003).
A final note on economic activity in Denizli concerns the main suppliers and customers of the cluster firms. Cotton processing factories mostly use cotton produced in the Denizli region. Yarn factories, on the other hand, use processed cotton not only frorn DenizJi and its environs but also from Isparta, Adana, Antalya. Gaziantep, Mugla and Aydin. Cotton yarn is also purchased from the provinces of Gaziantep, Adana, Antalya, Urfa, Hatay, Bursa and Kahramanmaras, and from abroad, particularly from Pakistan, Sri Lanka and Sudan. The fabrics used to produce towels and bathrobes come mainly from Denizli but also from Bursa, Istanbul, Izmir and Adana. The principal customers of the cotton processing factories are the yarn factories in Denizli and, to a lesser extent, lzmir. The main customers of the cotton yarn and fabric factories are in Denizli, too, although they also sell their products in Istanbul, Usak, Isparta, Ankara, Bursa and Izmir (Mutluer, 1995). Finally, the towel and bathrobe manufacturers export the great majority of their products to the EU and the United States. In other words, the final products of the cluster are exported and the remaining cluster firms are mostly organized to serve the needs of the producers of the leading product category.
Why Denizli? Why towels and bathrobes?
We turn now to the reasons for the concentration of towel and bathrobe production in Denizli. Why did this business concentrate in Denizli rather than anywhere else, and what brought about the specialization in towels and bathrobes rather than some other product category? The related topics of the perceived costs and benefits of clustering, and the self-reinforcing nature of clustering will be discussed separately in the subsequent sections.
As discussed earlier in this chapter, historical circumstances played a key part in the eventual emergence of the cluster in Denizli. The roots of the duster can be traced back to antiquity, when the region was host to major civilizations. This guaranteed an environment that was conducive to devel-opment and commercial exploitation of the high-quality local cotton. Although the region's fortunes waxed and waned over the centuries, invaluable experience in the production of textiles was accumulated. In the later decades of the twentieth century, development of the cluster was aided by central government measures (especially public investment in the 1970s; and the provision of incentives in the 1980s) and the support provided by local institutions such as the Chambers of Commerce and Industry, the Small and Medium-Sized Industry Association (KOSGEB) the Aegean Garment Producers' Association and Pammukkale University. The availability of capital accumulated through cooperatives and previous commercial activities, as well as remittances from Turkish guest workers in Germany who were keen to invest in their home town, was also beneficial to the growth of the cluster. Yet another factor was Denizli's favourable sociocultural environment, with strong entrepreneurial skills going hand in hand with financial relationships based on mutual trust and fellow citizenship. One example of this was private borrowing against the so-called 'Babadag Banknotes', or promissory notes, which were always paid on the due date. This traditional practice, although not as prevalent as it once was, still takes place in Denizli (Sengun, 1998). Borrowing money from family members and friends (rather than banks), and borrowing and buying machinery from other firms owned by relatives or friends were also common (Erendil, 1998), and of crucial import¬ance during the early years of an enterprise. Other types of support provided by the social network included knowledge and technology transfers (Eraydin, 2002a). The widespread use of subcontracting served to reinforce trust, espe¬cially with respect to vertical relations among cluster firms, as evidenced by the fact that firms still prefer to maintain long-term relations with trusted subcontractors.
Thus in answer to the question 'Why Denizli?, it can be argued that favourable historical circumstances, the availability of high-quality raw materials, the presence of talented craftspeople, a local demand for their products, and experience and know-how accumulated over the centuries led to the burgeoning of commercial activities in textile-related areas, aided by public investment, guest workers' remittances and state incentives. Without one final factor, however, the cluster might never have realized its potential: the entrepreneurial spirit of the people of Denizli. One of our interviewees pointed out that Denizli was not in fact an ideal location because, among other things, the quality of the water and the transportation services were poor, and therefore Izmir would have been a better location. However, given that the cluster emerged and developed in Denizli despite its locational disadvantages, it can be concluded that the factors identified above far outweighed the disadvantages. Besides, once the process of cluster formation begins, the dynamics of this process can turn the cluster into a self-reinforcing system. This will be discussed in a later section. Here it is sufficient to say that outside individuals, firms and institutions may be drawn into the locale after seeing the success of others and the benefits of concentration, thus feeding the reinforcement of the cluster.
Turning now to the second question posed at the beginning of this section - what brought about the specialization in towels and bathrobes? - ¬although textile production in the region (including towels) had deep historical roots, towel production became an important economic activity in Denizli only in the 1970s. This was mainly initiated by two events: the large-scale importation of looms used for towel production from Bursa in the 1950s, and the export success achieved by this product from the 1970s. One of the very first exporters of towels and bathrobes was a prominent entrepreneur from Babadag, A. Kadir Uslu, whose father was a merchant of cotton yarn. Uslu's willingness to take risks was such that when he wanted to import looms to produce towels his father was reluctant to help him financially, saying that he was 'a bit crazy'. Despite the lack of support from his relatives, he duly imported the looms and later went to Germany with a container load of towels. This venture was very successful, and he returned to Denizli with a significant number of orders, which motivated others to enter this area of activity.
The story of this entrepreneur raises the question of whether the emergence of the towel and bathrobe cluster in Denizli was purely a chance event. I would argue against this contention. It was no accident that Denizli came to specialize in towels and bathrobes. Rather Denizli entrepreneurs deliberately concentrated on the manufacture of these products and their export, aided by the economic infrastructure of Denizli, the main dimensions of which were a long history of textile production and the availability of major inputs, semiskilled workers and related/supporting industries. In other words the local business environment was conducive to such a development, and its potential was realized by the entrepreneurs of the region.
According to our interviewees, towels and bathrobes are amongst the easiest textile/clothing items to produce, which facilitated the entry of new comers and hence the enlargement of the cluster. Another factor was the international demand for these products. Arguably, local demand was an important factor only in the early stages of the cluster, although local customers are still demanding when it comes to the quality of textiles and clothing, for example people from Babadag can judge the quality of a product at first glance. However it is the cluster firms' export orientation that is now putting pressure on the firms to upgrade. Research and development are sometimes internationally driven, too: 'The impetus for new product devel¬opment comes from international customers. A customer from the USA, for instance, asked whether we could produce towels made of synthetic materials, and this triggered R& activity as wel1 as new investments in this area. The same is true of quality- and design-related issues.' Although international customers can be very demanding and ask for improvements in quality and service, the DenizIi producers have faced the challenge and specifically targeted difficult, demanding customers in stable developed markets.
Perceived costs and benefits of clustering
As mentioned previously, there are strong subcontracting relations among the cluster firms, enabling them to excel in their particular areas of specialization and to undertake work beyond the capacity of their firms. This ensures smooth and rapid production. With regard to intermediate tasks such as embroidery in particular, it is important for firms to have easy access to the required services and the degree of control afforded by geographic proximity. It is also important for the subcontractors to be competitive in their special areas of activity. As emphasized by one of our interviewees, 'if they are good, you do not need to worry about this part of the production process; you instead focus on the critical issues of your business such as quality and the marketing of your products'.
In a cluster environment, benefits accrue not only from vertical relations in the value chain but also from relations with related and supporting industries and competing firms. Apart from obvious advantages such as lower transportation costs, the presence of related and supporting industries also enables operations to run smoothly, just as in the case of subcontracting relations. According to one manager, 'It is reassuring to know that there are films nearby operating in complementary areas such as dyeing, machinery and packaging. I think this also encourages people to choose Denizli when they think about starting a new venture in this area.' Hence it is not surpris¬ing that suppliers of high-quality inputs, dyes, packaging and so on emerged in DenizIi in parallel with the development of the towel and bathrobe industry.
With regard to the benefits associated with the presence of competing firms, the cluster firms closely monitor each other in order to keep up with new developments and innovations. An interesting related observation is that there is an unspoken agreement on maintaining price parity on the main products of the cluster, and if one firm substantially reduces the price of, say, its bathrobes, this will immediately be brought to the attention of the other producers in Denizli. In this regard the cluster environment acts as a control mechanism to prevent destructive price competition, since it is clear that all producers will lose if such competition breaks out. One negative aspect of the colocation of rival firms is the immediate copying of ideas. However the managers interviewed seemed to agree that although this can cause some concern in the short term, it does not matter in the long term as it prompts firms to strive to achieve what others cannot do, stimulating a mutually reinforcing process of upgrading and the pursuit of best practice. All this is easier and more rapid in a cluster context, and is facilitated by the existence of social networks and a sense fellow townsmanship that aid the flow of information.
Thus information spillovers enhance the self-reinforcing process that feeds innovation in the cluster. As mentioned earlier, new information is obtained from various sources, including trade fairs, catalogues, subcontr¬actors, customers and the Denizli Chambers of Commerce and Industry. Moreover members of industry associations meet regularly to discuss their problems and find remedies for them. Information on markets, strategies, customers and products are especially important - even more so than tech¬nological information, according to some of the interviewees. One inter¬viewee mentioned that a few duster firms prefer to register their exports not in Denizli but in Izmir or Antalya to avoid the leakage of information on key customers and products.
In summary, the case of the towel and bathrobe cluster in Denizli shows that a number of benefits can be derived from clustering. Apart from being convenient for customers (see the next section), control over subcontractors is much easier in a cluster context, ensuring smooth and rapid production. Social relations and relations of trust among the major actors in the region also contribute in this respect, as well as promoting a better flow of information. Further benefits are obtained from related and supporting industries, and even competitors. The only problems with the proximity of competitors are the rapid imitation of key strategies and the possibility of a price war, although the latter is unlikely as all parties would incur high costs. Similarly the former, although annoying for innovative firms in the short term, can produce fruitful results in the long term if it encourages firms to develop more sustainable sources of competitive advantage.
The self-reinforcing nature of clustering
Once a cluster is formed and successfully functioning, a positive feedback mechanism might come into operation, turning the cluster into a magnet that attracts other firms in the same or complementary lines of business. This happened to the Denizli towel and bathrobe cluster in the 1990s, when the city became well known in the world market. As towel production and export only began in earnest in the 1970s, cluster formation in this particular case took about 20 years.
To illustrate the benefits of the self-reinforcing system that comes to operate in a well-functioning cluster such as the one in Denizli, consider the example given by one of our interviewees of a hypothetical entrepreneur in Nazilli (an Aegean town about 80 kilometres from Denizli), who wishes to establish a firm specializing in towel and bathrobe production. Nazilli also has a background in textiles and a similar, if not better, infrastructure and workforce than Denizli Let us assume that this person has already imported the necessary machinery. He is likely to have problems right from the start of his endeavour; that is, setting up the right layout in the factory, for which he will probably need to employ a master craftsman from Denizli. Having sorted this out, the next step will be to buy cotton yarn. In order to do this at favourable terms he will have to go to Denizli, where the numerous merchants located in the City control this line of business. These merchants know all their DenizIi customers very well, as well as their families. It will take some time for the entrepreneur to establish such a close relationship with the yarn merchants and gain their trust. Interpersonal relations are very important in securing a competitive price, good-quality products and timely delivery. Another problem for the entrepreneur is that textile machines break down frequently, again requiring him to go to DenizIi to find someone to repair them. He will also have to go to Denizli if he has to subcontract some of his factory's operations. Moreover he will have to arrange the trans¬portation of all inputs and products to and from Denizli. Assuming that he manages to overcome these obstacles, the next question he will face is how to market his towels and bathrobes in Nazilli. As pointed out by the interviewees, the Denizli industrial zone also functions as a market, and customers visit many firms before finalizing their purchasing decisions. The geographic concentra¬tion of the firms makes it convenient for the customers to visit them rather than the other way round. In addition the related services are organized according to the needs of the cluster. For instance the personnel working in the banking sector in Denizli have detailed knowledge of the specific problems of the manufacturers operating there. In short, it will be virtually impossible for the hypothetical entrepreneur in Nazilli to keep up with the pace of business in Denizli and compete with its firms.
In light of the above it is not surprising that none of the interviewees intended to leave his home town and change the location of his factory. Frequently heard statements in this regard were 'We know how to do business here', 'We can follow the latest developments in our field here', and 'It is good to know that there are other enterprises in Denizli you can turn to if you need help and support'. These statements are in conformity with the finding of a survey by Mutluer (1995) that the locational advantages offered by Denizli were the principal reasons why 65 per cent of entrepreneurs decided to establish their enterprises there. A key factor in the growth of the cluster is new business formation, which usually takes place via spin-offs. The process works as follows. The growth of an enterprise provides its partners (who are usually relatives or fellow townsmen) with the financial means to found a business of their own, and well established contacts give them the confidence they need to engage in business operations by themselves. They are also familiar with the particular way of doing business in the locale, which is by no means guaranteed to be the same in another location. It should be noted that although spin-off firms tend to continue their contact with the original firm in the beginning, a delicate dynamic arises as the former may well become a competitor of the latter (Eraydin, 2002a).
With regard to the relative importance of the factors that contribute to the self-reinforcing nature of the cluster, external economies appear to be the leading ones. Given that many other locales in Anatolia have a long history of textile production, as well as similar infrastructural and factor conditions, why did external economies in towel and bathrobe production emerge and turn into a self-reinforcing system in Denizli rather than some¬where else? There are a number of possible answers to this, some of which were outlined above when discussing the reasons for geographic concentration. If we consider one possible reason, entrepreneurship, the question then is, why are Denizli entrepreneurs better than entrepreneurs elsewhere? This, it can be argued, is due to the characteristics of the local business environ¬ment, competitive pressure in particular, which force people to sharpen their entrepreneurial skills, so the challenge now is to identify these characteristics. This can only be done by combining the analyses of all the case studies in this book, so the task will be left until Chapter 8. Before closing this section, it should be emphasized that a combination of factors - ranging from historical circumstances and factor conditions to the nature of competition and the sociocultural infrastructure - needs to be considered when trying to understand the underlying reasons for the emer-gence and subsequent development of this so-called 'a la Porter' cluster in full (Kucuker, 1998, p. 10). From a theoretical point of view, the towel and bathrobe cluster in Denizli resembles a number of geographic clusters in Italy in respect of the largely spontaneous nature of its development, its specialization and its strong export orientation. An entrepreneurial outlook and the rapid pace of new business formation, especially via spin-offs, are other noteworthy commonalities. The exact nature of the link between the self-reinforcing dynamic attained by the cluster over the years and the sustainability of its competitiveness is of special interest when attempting to understand the relation between clustering and competitiveness more generally, and it is to this that we turn to in the final section of this chapter.
Concluding remarks and future prospects
The analysis in this chapter has highlighted the important part played by historical circumstances, favourable factor conditions, related and supporting industries and a favourable context for firm strategy in the success of the Denizli cluster. Specifically, the low cost of labour, relatively cheap but high-quality raw materials (especially cotton), government incentives, the entrepreneurial ability of Denizli businessmen and a clear focus on quality have been instrumental in this success. Although some deficiencies remain, such as the lack of qualified personnel and capital and the need for infra¬structural improvement, Denizli has been one of the world's leading centres of towel and bathrobe production since the early 1990s. Its contribution to the Turkish economy is also noteworthy. For instance each year around 10 textile firms from Denizli are listed amongst the top 500 Turkish firms, ranked according to their sales revenues by the Istanbul Chamber of Commerce. In fact Denizli has been chosen as a model of local industrial growth by the State Planning Organization (TBMM, 2001a).
The success of the Denizli firms has attracted further investment to the area, including banking facilities. As a result the cluster firms, which previously mainly relied on their own financial resources (and were proud of doing so), have gained access to external means of financing. This, however, caused severe financial - especially liquidity - problems for the firms in 2001, when the Turkish economy in general and banking sector in particular was hit by a major crisis (ibid.). There was perhaps no better test of the resilience of the cluster than a crisis of this scale, a 'test' that it managed to survive.
In fact the February 2001 crisis came on top of a series of crises that hit the Turkish economy in the 1990s and badly affected many industries across the country. Specifically, in the early 1990s the Gulf crisis had a negative effect on profit margins. Then in 1994 the Turkish financial crisis had a particu¬lar impact on output, employment and domestic sales, and resulted in an increase in the cost of financing (Temel et al., 2002). This was followed by the Asian and Russian crises of the late 1990s. In response to all these crises the cluster firms developed the 'Free Trade Zone' and 'Transformation 2000' projects. The purpose of the former was to provide tax advantages to firms, and the second was aimed at improving their level of technology (Eraydin, 2002a). According to our interviewees, compared with many other regions of Turkey, Denizli was less damaged by the crises, mainly because the cluster was highly export-oriented and its exports were directed towards stable, developed market economies. Apparently they learnt the art of crisis management very well, which was a must under the circumstances. One of the interviewees stated that he had been working as a manager for seven years and could not remember a single year going by without a serious national economic crisis.
Any discussion of the present situation of and future prospects for the cluster must include the strategies followed by the firms. There is an under¬standing amongst the firms that any strategy based on cost advantages such as cheap labour and low-cost inputs is not sustainable in the long term. According to the interviewees, the advantages of low-cost labour and good-quality but cheap raw materials were enough to secure a satisfactory export performance for the cluster firms until the end of the 1980s. Thereafter, however, new competitors emerged, especially in Pakistan, India and China, which prompted the Denizli firms to put more emphasis on quality as the price-sensitive segments were gradually lost to them. A major factor in the realization of this shift was the significant amount of capital that had been accumulated .in the 1980s, when mostly standard products had been produced. This capital was used to finance new investments, especially in technology, thus enabling the cluster firms to improve the quality of their products. As touched on earlier, demanding international customers also played a key role in this respect, as evidenced by the following statement by an interviewee: 'We have come to the point where we cannot compromise with quality. This we have learnt from our international customers. . .We do not run away from their tough requirements and stringent standards. When a customer demanded fireproof baby clothing made from terry fabrics, for instance, we worked on it and brought it about.' The interviewees argued that they had improved the quality of their products to such an extent that producers in other regions of Turkey were finding it difficult to keep pace. They had also done much to improve service quality.
The cluster firms are aware that their future prospects will not be bright unless they find more sustainable sources of competitive advantage. To this end they have not only improved their product quality but also increased their product range. In strategic terms there are two possible courses that the firms can take: develop a brand of their own, or become producers for global brands. The former is the subject of considerable debate amongst the cluster participants. Some of the interviewees were of the opinion that developing a brand was a must, while others considered it to be an unneces¬sary and unrealistic target. One of the managers in favour of developing a brand offered the following illustration of his point: 'We sell a bathrobe to the USA for $30, and it ends up in a luxury store in New York where it is sold for $120.' To capture some of the $90 margin, he argued, it was essential to build up a brand name. In fact some leading cluster firms have already made progress with this. There are, however, problems with marketing and devel¬oping professional strategies, which are in turn related to the limited availa¬bility of qualified personnel and finance: 'Building a brand is expensive, requires time and finance. Most importantly, however, it requires an aes¬thetic, an ability to develop a style, a cultural base, and an ability and will¬ingness to follow developments in the world.' The major challenge, in other words, is to overcome the limitations imposed by the current resource base of many firms, especially with regard to qualified personnel.
A related concern is the risk involved in building a brand, and to date only a few firms have been successful in this endeavour. One of our inter¬viewees argued that this was a matter of specialization: 'We tried to develop a brand and we went bankrupt. As Calvin Klein said, retailers should not attempt to be producers, and vice versa.' In fact, because the sector is domi¬nated by a large number of small firms it is unrealistic to expect that all firms will be able to develop widely known brands. According to one interviewee, it will take some time but it can be expected that about five or six strong brands will be established, and that will be enough to promote the image of Denizli in the world market. Another interviewee suggested that instead of individual firms developing brands of their own, it might be a better idea to promote the name of Denizli in the world market. The latter course and attempts by individual firms to build a brand of their own are not mutually exclusive, unless what is implied is the development of a single brand for the entire network of firms.
It is often claimed in the literature that clusters in developing countries are locked into serving as Iow-cost subcontractors to firms in developed economies, which is not always desirable. Moreover it is difficult for devel¬oping-country subcontracting firms to sustain their relative positions in the world market since they can easily be displaced by others offering even lower prices. The need to develop more sustainable sources of competitive advantage is also imposed by the changing circumstances in the business environment. Only if this is achieved can a subcontractor be an indispensable associate of a global firm. The following statement by an interviewee is informative in this respect: 'If it turns into a relation based on mutual dependence, subcontracting is probably sustainable', but this is only possible if subcontractors continuously improve their service and become irreplaceable. 'We should avoid doing too simple things', he continued, 'we shall only have a chance in the future if we can produce textile items that low-price competitors cannot.'
Yet another issue that should be considered is that there is concern about the high degree of specialization in the Denizli economy, that is, it is dependent on the textile industry in general and the towel and bathrobe cluster in particular. Some organizations consider that the dominance of the towel and bathrobe cluster constitutes a serious constraint on the local eco¬nomy and warn that the lack of variety is risky and might create problems in the long term (DTO, 2003; IAV, 1997). One suggestion is to encourage investment in the other sectors in which Denizli has a strong potential: food and beverages, iron and steel, marble and tourism. If a crisis should affect the cluster it would be reassuring to know that there were other areas to turn to, but our analysis shows that the benefits of clustering are significant. One should also be careful about sweeping generalizations such as the argument that the sector should be abolished since it is old.fashioned and mature and therefore has no future. This argument is raised for the textiles/apparel sector more generally, which is one of the leading sectors in the Turkish economy. Italy offers an illustrative example of a country that has achieved a considerable degree of development as well as retaining an impetus for further growth despite its specialization in 'old fashioned' or 'mature' industries. It is how these industries compete - that is, their strategy - that really matters (Porter, 1990).
Some researchers (for example Erendil, 1998) suggest that a likely conse¬quence of the recent tendencies observed in Denizli could be a gradual loss of economies of agglomeration since, with the sector becoming more differ¬entiated and segmented, interfirm relations will change and the necessity of collective action and collaboration will diminish. This reflects an incomplete evaluation of the reasons for geographic concentration: firms do not only agglomerate in order to cooperate. Rather there are gains to be had from being located close to competitors and sharing the same business environment, even if competing firms are unable and/or unwilling to cooperate when organizing production. Besides the available evidence suggests that the colo¬cation of rivals might lead to differentiation but will not inevitably result in the dissolution of clusters (Baum and Haveman, 1997).
In summary, circumstances are changing and it remains to be seen how the towel and bathrobe cluster in Denizli will react to recent challenges. Denizli has now reached the stage that can be observed in many other clusters in the world: having confronted the consequences of its initial attempts to integrate itself into foreign markets, it now faces the challenge of overcoming the negative factors associated with less developed countries and integrating itself into international markets by means of higher value-added products (Eraydin, 1997). Although Denizli has been described as a typical example of Porter-style clustering (Kucuker, 1998), it will have to find more sustainable sources of competitive advantage if it is to become a fully functioning, archetypal Porter cluster. This will be of determining importance in defining its relations with global producers as a local production centre. Transformation is necessary, but it will not happen automatically. Rather it will depend on firms' willingness to change, the availability of qualified human resources and local (private and public) institutions; that is, following the right strategies in a business environment that is conducive to improvement. The analysis provided in this chapter has shown that there are reasons to be optimistic.