Clusters and Competitive Advantage_7 pdf

Currently there are three small industrial estates containing thousands ofsmall and medium-sized enterprises, mostly in the machinery, metal-working,transportation equipment and furnitur

government has been even more meagre (Eraydin, 2002a), as in the casesexamined in the previous chapters

Reasons for geographic concentration

Initially the industrial establishments in Gaziantep were concentrated in andaround the city centre In the 1950s the Sehreküstü district and Nizip Streetand its environs became industrial spaces, and in the 1970s KÜSGET andÖrnek Sanayi Sitesi industrial estates18 were established on the periphery of thecity Currently there are three small industrial estates containing thousands ofsmall and medium-sized enterprises, mostly in the machinery, metal-working,transportation equipment and furniture industries, and three organizedindustrial zones for firms in the textiles, machinery, chemicals/plastics andfood industries; that is, the leading businesses in the Gaziantep economy

In the city centre, Nizip Avenue and its environs are still popular areas forindustrial establishments.19 Another central area, the Ünaldi-Sehreküstü district,

is home to small and medium-sized lentil processing facilities and carpetfactories Interestingly, according to our interviewees carpet firms typicallystart their operations with an office in the city centre and manufacturingfacilities in the Ünaldi-Sehreküstü district, but then move to KÜSGET andfinally to one of the organized zones In other words, Ünaldi-Sehreküstüand KÜSGET are seen as stepping-stones to the organized zone.20

The emergence of the cluster

Before proceeding to the particular circumstances that led to the emergence

of the carpet cluster in Gaziantep, it might be helpful to consider why woven carpet production mostly takes place in western Turkey in generaland central Anatolian cities in particular According to Ayata (1987), this isrelated to the nature of the land Although land is not scarce in centralAnatolia, few types of crop can be cultivated there (mainly wheat andbarley), and in past days this left the peasants with plenty of spare time toengage in a variety of handicrafts, including carpet weaving.21 Thus the nature

hand-of agricultural production gave rise to the practice hand-of carpet weaving in centralAnatolia

In the case of eastern Anatolia, however, there are few historical records ofcarpet weaving as an important economic activity, apart from some production

of kilims Arguably, the fact that kilims were produced in Gaziantep facilitatedthe transition to the carpet sector According to one of our interviewees, themain catalyst for the emergence of the cluster was the growth of cross-borderimports of silk carpets for decorative purposes (usually for wall hangingsand bedspreads) These were very popular, not only in the immediate regionbut also in the rest of the country The interviewee thought that kilim weavershad begun to produce such carpets in Gaziantep in an attempt to capitalize

on this trade However they were unsuccessful and decided to concentrate

The Carpet Cluster in Gaziantep 129

on the production of regular carpets for floors, thus ‘building on their mulated experience in kilim manufacturing’ Many of the intervieweesunderlined the importance of the existence of a textile sector in Gaziantep

accu-in general and the highly developed yarn and textile-machaccu-inery sectors accu-inparticular Gaziantep’s master craftsmen had no difficulty copying loomsand other carpet-manufacturing machinery imported from other parts ofTurkey and abroad, so the transition to machine-woven carpet productionwas quite smooth Another interviewee argued that a change in demandpatterns had brought about the transition to carpet production, in that ‘ifthere is demand for a product and one firm succeeds in fulfilling this demand,others follow suit’ What we can infer from all this is that the emergence ofthe machine-woven carpet cluster in Gaziantep can be explained by acombination of factors: the existence of a number of related and supportingindustries (the yarn and machinery sectors in particular), demand patterns,and ‘historical accidents’ such as the success of imported silk carpets This explanation resembles Krugman’s (1991a) account of the emergence

of the carpet cluster in Dalton in the United States, which is presented as atypical example of a historical accident triggering the beginnings of a cluster

It is also very informative in respect of cluster formation in general, especiallywhen considered together with the emergence of the towel and bathrobecluster in Denizli In both cases there was a long history of textile productionbut the later specialization in a particular product group (towels and bath-robes in Denizli and machine-woven carpets in Gaziantep) was verymuch the result of a historical accident, the local availability of keyresources and an entrepreneurial spirit Successful ventures were noted byother potential entrepreneurs in the region who were motivated to replicatethe success of their fellow townsmen This in turn fed new business formation,initiating the self-reinforcing cycle described in Chapter 5 When links wereestablished between the carpet firms that formed the core of the clusterand related/supporting industries and institutions, the benefits associatedwith external economies strengthened the self-reinforcing nature of clustering.Therefore, a phenomenon that looks like a purely chance event might infact be explained by the specific attributes of the local environment, andaccidents of history are successfully capitalized only when the environment

is favourable

Amongst the factors that have helped Gaziantep’s carpet producers toprosper, the part played by the yarn and machinery sectors is of specialimportance As mentioned earlier, the yarn industry is one of the oldesttextile-related operations in Gaziantep, and it is very competitive Similarlythe machinery cluster, which is also a historically important field of activity

in the city, has developed and specialized over time, reflecting the needs ofthe sectors that dominate the economy Cooperative networks have also had

a role in the development of the cluster, although not as much as in Denizli.For instance firms help each other in times of crisis Strong ties with family

and friends, on the other hand, have been particularly influential in theprocess of new business formation For example people who originally camefrom Besni (a town to the north-east of Gaziantep), who have a good reputationand prominent position in the Gaziantep carpet sector, have helped many

of their relatives and friends to enter the business by providing them withguidance on the particular way of running a business in the cluster Having

a mentor to ask for advice not only facilitates entry but also improves theprospect of long-term success

The benefits of clustering

When asked about the advantages of Gaziantep as a business location, anentrepreneur from Besni explained that the lack of opportunities in Besnihad forced him to move to Gaziantep, which offered more in the way ofindustrial pursuits Since he was familiar with textiles and kilim production

he had channelled his efforts in that direction Relatives and friends with

an established position in the city had helped him to set up his business

‘It is good to know people working in the same area as you are You ask forideas, where to buy yarn, for instance It is much easier to run your businesswhen you have good contacts.’ Other advantages are the city’s good infrastruc-ture, qualified workers, key location, accumulated know-how, including how

to conduct international business activities, and the presence of related andsupporting industries and institutions.22 The benefits associated with this envir-onment are significant For instance one interviewee mentioned that most

of the entrepreneurs who had bought the necessary machinery in Gaziantepand started a new carpet-weaving business in another city had failed: ‘State-of-the-art machinery alone is not enough for the success of a business Ifyou try to run the same business in Ankara, for instance, you might find itdifficult If a machine breaks down you may not even find someone torepair it.’ Another example relates to a carpet firm in Çankiri in centralAnatolia: ‘The manager of this firm has to buy yarn from here He calls usevery day to keep up with the latest developments in the industry Heregrets that he did not establish the firm in Gaziantep.’ As a result, in thewords of another interviewee, ‘the rationale that there are too many firms here,

so I will go and establish my firm somewhere else’ does not work becausethe quality of the local business environment is paramount

There is also a concentration of firms in unrelated businesses, including,pasta and footwear One reason for this, according to the interviewees, is thatthe local people have an entrepreneurial spirit: ‘If Gaziantepians smell anopportunity, they enter a business impulsively’.23 Apparently they have thecourage to take risks – ‘If they have one million lira they do not hesitate toinvest in a business that requires two million’ – because ‘an entrepreneurialmindset’ is instilled in them as children

There are some general benefits of clustering that are largely independent

of the location in question, such as the proximity of related businesses, which

The Carpet Cluster in Gaziantep 131ensures the smooth functioning of an enterprise As mentioned earlier, allkinds of yarn are readily available in Gaziantep, and the exact quantityrequired can be bought at favourable terms and very quickly This has costadvantages (especially in terms of transportation and storage costs), it ispractical and it facilitates information flows among cluster participants How-ever opinions differ with regard to the benefits of proximity to rival firms.Some of the interviewees pointed to the danger of destructive price competitionand the rapid imitation of new designs and best practices, whereas othersemphasized the dynamism this brought to the sector by encouragingparticiants to improve their competitive advantages Nonetheless there wasconcensus that the colocation of rivals was advantageous for customers As

in the case of Denizli, it is the customers who visit the carpet firms inGaziantep, rather than the other way round One interviewee likened this tobuying tomatoes in the local market, where there are numerous sellers andthus more choice In fact the same logic applies to carpet manufacturers asthey are customers of the yarn and machinery firms: ‘We have a chance tochoose among the different alternatives .available in Gaziantep We can getinformation on all the details and are able to make an educated decision.’Another interviewee thought that colocation aided publicity: ‘It is difficultfor a single firm to be known in the world carpet industry But as Gaziantep

is well-known, you can be known as a carpet firm from Gaziantep.’

Returning to the divided opinions on whether colocation resulted indynamism or posed the threat of destructive price competition, the inter-viewees stressed that price competition was more likely to happen when therewas stagnation in the market Otherwise, as discussed in Chapter 5, the clusterenvironment might work against such competition, since the likely damagewould be evident to the local actors and there might be social sanctions aswell (see Chapter 8 for a detailed discussion of this issue)

Another aspect of the proximity of rival firms is the flow of information

In the Gaziantep carpet cluster, information flows come mainly fromcustomers, designers and suppliers ‘Customers are very demanding in thissector They force you to follow the most recent trends.’ As many of thesmaller carpet manufacturers work with the same designers, key information

on the most recent trends is quickly diffused The same holds for yarnmanufacturers and merchants Moreover factories are located within a step

of each other, managers and workers know each other, and the state of thebusiness is a popular topic of conversation As one interviewee put it,information flow is so speedy that ‘it would not be more rapid if it werebroadcast on TV’ One further implication of this concerns the fact thatpreserving one’s reputation and behaving according to the norms prevailing

in the cluster are of vital importance It follows that any news of damage toreputation or deviation from the norms will disseminate rapidly as well

In the following section we shift our focus to another of the major concerns

of this study: the likelihood of the cluster sustaining its success, which

requires an analysis of the changes that have taken place in the environmentand how the cluster has responded to them

Concluding remarks and future prospects

Our analysis of the history of the cluster has revealed that although textileproduction has a long history in Gaziantep it was not until the 1970s thatthe production of machine-woven carpets began in Gaziantep and the clusteremerged The second half of the 1980s witnessed the internationalization ofthe cluster, and in the 1990s it transformed itself by upgrading its technologyand improving the quality of its products

Meanwhile important developments were taking place in the externalenvironment, some of which had a lasting impact on the cluster For instancethe Gulf crisis in 1991 affected the Gaziantep economy much more thanelsewhere in Turkey because the city’s significant trade relations with theMiddle East in general and Iraq in particular were severely damaged (Eraydin,2002a) The cluster firms were, however, lucky to have several newly emergingmarkets that they could turn to in the meantime, the Russian Federationand the CIS in particular The Gulf crisis was followed by a series of financialcrises in Turkey in the 1990s and early 2000s, which worsened their exist-ing problems (such as the high cost of financing) and badly affected theirperformance in terms of sales, profitability, investment and employment.Despite these setbacks the firms have maintained their key advantages: low-cost

labour and high-quality products (Temel et al., 2002)

The way in which the cluster firms have reacted to changes in externalcircumstances over time is reflected in their strategies As mentioned whendiscussing the context for firm strategy, in its early stages of development theGaziantep carpet sector had a rather poor reputation for quality However,this is no longer the case, as since the 1990s firms have been forced to givepriority to quality in the face of increased competition, especially fromBelgium Although Istanbul, Kayseri and Nigde have some presence in themachine-woven carpet industry, they have ceased to be key competitors:

‘These places are different from Gaziantep; they have a smaller number of verybig, integrated firms’, one interviewee explained The interviewees spokepositively about the strong competition from Belgium, whose entry into the CISmarket, for instance, had forced them to pay more attention to quality in thatmarket as well Gaziantep has a cost advantage over Belgium and because ofthis some Belgian carpet firms are planning to invest in Gaziantep; in factone of the leading yarn factories in Gaziantep already has a Belgian partner

In the mid-1990s new regulations on patents also triggered a change ofstrategy amongst the cluster firms Before the proper enforcement of thepatent law, copying was widespread and some firms even used the exactbrand names of companies from which they had copied designs Followingthe implementation of patent law, some firms were required to pay huge

The Carpet Cluster in Gaziantep 133fines and others went bankrupt One average, however, the firms adapted tothe new circumstances very quickly and made original design a principalconstituent of their strategy In a very short period of time the cluster wastransformed from one that stole designs from the leading firms in the industry

to one that invested in, developed and exported creative designs

Although these developments have generally been welcomed by thecluster firms since it is recognized that design and quality are the real com-petitive factors that differentiate firms in an environment where inputs areavailable to all at more or less at the same terms, serious attempts to buildbrands have yet to take place in Gaziantep (a notable exception is Merinos)

As noted earlier, the cluster firms instead capitalize on the good reputation

of Turkish carpets in the world market

Some scholars (for example Temel et al., 2002) are sceptical about the

competitive prospects of the cluster, given that developing and neighbouringcountries are key export destinations for the cluster’s products There arealso problems with marketing and after-sales service (Eraydin, 2002a) Thefirms themselves, however, are very optimistic about their future prospectsand expect Gaziantep to replace Belgium as the world’s leading centre formachine-woven carpet production within a couple of years, given thatBelgian manufacturers have found it difficult to compete with Gaziantepacross a broad range of products and have tended to narrow their range Before closing this chapter it should be reiterated that the Gaziantepeconomy does not exclusively rely on one sector, unlike that of Denizli It isenvisaged that the structure of the economy will remain the same (Özsabun-

cuoglu et al., 1999) and that the food (flour, pasta, meat and related products,

milk and related products, alcoholic beverages, olive oil), textiles (yarn, cottonfabrics, woollen fabrics, carpets), leather (leather processing, footwear), soapand miscellaneous machinery sectors will continue to prosper Gaziantep’ssuccess is of symbolic importance, given the low level of developmentattained by the other provinces in the region In this regard Gaziantep isconsidered an important regional centre, and there might well be spillovereffects on neighbouring provinces, as can already be observed in the case ofKahramanmaras

The analysis of the Gaziantep carpet cluster reinforces the main conclusions

of the analysis of the towel and bathrobe cluster in Denizli In both casesthe quality of the local environment is of determining importance in res-pect of the nature and sustainability of competitive advantage Moreoverboth clusters have accumulated considerable experience in the general field ofactivity, coupled with the local availability of resources and an entrepreneurialspirit However competitive pressure is more pronounced in Gaziantep,and firms in the latter are much more reluctant to collaborate Competitivepressure – both national and international – and how firms and entrepreneursreact to this pressure appear to be key factors in the dynamism of theGaziantep cluster and the sustainability of its competitiveness In fact the

latter is one of the factors that differentiates the cluster from the furniturecluster in Ankara, the other being that Gaziantep has accumulated consider-able knowledge and experience of conducting international business, anadvantage that the cluster in Ankara lacks.

in Istanbul, which has gone through a strategic transformation over the pasttwo decades Interestingly, the change in the strategic orientation of firmshas been accompanied by changes in the location of the industry withinIstanbul, with the production and sales functions being concentrated indifferent locales Specifically, the tanneries and factories are located inTuzla, and firms that target the upper end of the market have retail outlets

in Zeytinburnu while those which target the lower end of the market utilizethe marketing channels in Laleli (Figure 7.1) The analysis of this cluster willtherefore provide us with new insights into the evolution of competitiveadvantage in clusters and enable us to comment on the dynamics associatedwith the location of different activities in the value chain Additionally, ananalysis of the informal Laleli market will provide us with the opportunity

to highlight a relatively understudied topic in the literature on clusters; that

is, the fact that transnational linkages are formed not only by multinationalenterprises but also by small entrepreneurs (Yükseker, 2003), and that informaltransnational activities might also cluster in space

Since the emergence of informal cross-border trade from Laleli in the late1980s after the collapse of the Eastern bloc, official statistics have failed toreflect the true export performance of the industry It is estimated thatinformal trade in leather garments amounts to $2.5 billion, while registeredexports of leather garments stand at $300 million (SPO, 2000) Istanbulaccounts for more than 80 per cent of total Turkish leather exports (ITKIB,2003), and it can safely be argued that this figure is much higher as mostinformal trade is conducted in Istanbul According to official figures, Turkey’scurrent world export share in leather garments (SITC 84811) is around

10 per cent Thus even if only registered exports are taken into account, it isclear that Turkey is among the top exporters in this product category,together with China, India and Italy (ITC, 2002) The leading registered

customers of Turkish leather garments are Germany, France, the RussianFederation, Spain, the United Kingdom, the United States, Austria, Switzerland,Belgium, Luxemburg and the Netherlands, while the main destinations forinformal sales are the Russian Federation, the CIS and Eastern Europe Turkeyalso imports some leather garments (valued at about $12 million) fromIndia, China, Germany, Pakistan and Italy (SPO, 2000, p 79)

Origins and historical development

Leather making in Anatolia can be traced back to the Hittites (1400 BC), and

it is known that the Asian Turks produced fine leather products at least 2400years ago (Yelmen, 1992) Istanbul’s emergence as a leading centre of leathermaking, however, is mostly associated with the early Ottoman era.2 Ottomanrule started in Istanbul in 1453 The city became the capital of the OttomanEmpire a few years later and stayed so until the empire collapsed in the earlytwentieth century The conqueror, Sultan Mehmet II, paid attention to theeconomic development of the city in general and that of the leather industry

in particular It was Sultan Mehmet who encouraged the founding of a leatherindustry in the district of Kazliçesme (Figure 7.1), which remained the heart

of the industry for more than 500 years, until its move to Tuzla in the early1990s (Yelmen, 1992; Kuban, 1996)

There were about 360 tanneries in Kazliçesme in those early years,surrounded by workshops that specialized in the manufacture of productsused in leather processing (Kömürcüyan, 1952; Yelmen, 1992) During theexpansion of the Ottoman Empire, considerable progress was made by the

KazliçesmeZeytinburnuBLACK SEA

The Leather Clothing Cluster in Istanbul 137

tanning industry In fact the sixteenth and seventeenth centuries were thegolden years of the Ottoman leather industry,3 which made all the shoesand other leatherwear required by the empire’s far-flung armies (Union ofChambers of Commerce, Industry and Commodity Exchanges of Turkey,

1958, p 3) Leather was used not only for footwear but also for clothing,saddlery, upholstery and bookbinding (Faroqhi, 2000; Mantran, 2001).Some items were produced especially for the Ottoman aristocracy, such asimperial gowns for sultans, leather saddlebags and leather-bound Korans Inthe sixteenth century, leather making was also an important pursuit inEdirne, Kayseri, Ankara, Bursa, Konya, Tokat and Diyarbakir (Figure 7.2), butalthough the ‘various tanneries scattered all over Anatolia were in continuousoperation to supply the needs of the vast territories of the Turkish Empire’(Union of Chambers of Commerce, Industry and Commodity Exchanges

of Turkey, 1958, p 2), Istanbul was clearly the principal centre of leather tradeand production (Faroqhi, 2000) While Istanbul’s tanneries were still con-centrated in Kazliçesme in the eighteenth century, other processing unitshad sprung up in Eyüp, Tophane, Üsküdar and Kasimpasa (Yelmen, 1992) During this golden era, leather produced in Istanbul was exported to someEuropean countries In fact historical records reveal that the technologicalaptitude of the industry in this era was supreme (Adnan, 1935), and thatsome Istanbul craftsmen gained international recognition for their skills(Faroqhi, 2000) Moreover it was not uncommon for foreigners to go to Turkey

to learn the methods and secrets of leather making, and in the eighteenthcentury a French doctor and an English citizen were rewarded for their success

Figure 7.2 Leading centres of leather production and trade, Turkey (current borders), sixteenth and seventeenth centuries

In the nineteenth century, however, the Ottoman leather industry started

to decline, together with the empire itself The gradual recession was

a consequence of the reluctance of the industry to adopt new techniquesand the absence of training (Union of Chambers of Commerce, Industryand Commodity Exchanges of Turkey, 1958), as well as the 1838 Ottoman–British Commercial Agreement, which brought about a vast increase inleather imports (Faroqhi, 2000) Another noteworthy event was the purchase

of a leather-processing unit (named Debbaghane-i Amire) in Beykoz (Istanbul)

by the government in 1810 to manufacture leather products for the Ottomanarmy In 1882 a shoe factory was established next to this factory In theRepublican period, this was taken over by a state-owned enterprise, Sümerbank,

to produce reasonably priced, good-quality shoes, suitcases and other items(SPO, 2000).5

In the early years of the Turkish Republic, before state planning was duced, leather making maintained the nature of a craft activity,6 a notableexception being the activities of the giant Sümerbank However in the 1950s,when the country’s economic policies were relatively liberal, there was

intro-a grintro-aduintro-al development of the industry This continued in the 1960s intro-and1970s, which were characterized by a planned approach to development The industry accounted for a large proportion of employment in Istanbul

in the 1950s and 1960s, followed by Balikesir, Bursa, Gaziantep and Izmir(mechanized), and Denizli, Maras and Usak (non-mechanized) (Union ofChambers of Commerce, Industry and Commodity Exchanges of Turkey,1958; Sümer, 1969).7 Over time the city of Istanbul expanded into peripheraldistricts and Kazliçesme, the heart of the Istanbul leather industry, became

a central district, creating problems for the producers located there Forinstance it was difficult for them to enlarge their operations, and the pollutionthey created affected nearby residential areas (Canyoldas, 1969) To resolvethese problems the Tuzla organized industrial zone project was launched in

1986 and completed in 1992 The move from Kazliçesme to Tuzla was, ever, poorly handled and resulted in losses in production (SPO, 2000).The government’s export-oriented economic policies in the 1980s benefitedthe industry enormously in that producers made good use of the exportincentives and tax rebates provided The leather clothing industry in particularperformed impressively in the world market, its world export share exceeding

how-20 per cent in the mid-1980s From the second half of the 1980s, however,there was a decline in its relative position To explore the reasons for thisrapid increase and subsequent decline we need to investigate the industry’ssources of competitive advantage, which is the main subject of the follow-ing section Before closing this section, a final note on the Istanbul economy

is needed to complete the analysis of the context in which the leather-clothingcluster emerged and flourished Table 7.1 provides an overview of employ-ment in Istanbul by economic activity It is clear that the Istanbul economydoes not depend on a narrow range of activities On the contrary, being the

Table 7.1 Employment in Istanbul, by economic activity 1

ISIC Sector

Location quotient

Share of national employment

6599 Other financial intermediation n.e.c

(not elsewhere classified)

1730 Manufacture of knitted and

crocheted fabrics and articles

3.2861 80.44

1912 Manufacture of luggage, handbags etc 2.9101 71.24

3330 Manufacture of watches and clocks 2.9039 71.08

3130 Manufacture of insulated wire and

6412 Activities of other transport agencies 2.7631 67.64

7414 Business and management

consultancy activities

2.7542 67.42

5131 Wholesale of textiles, clothing and