GaWC Research Bulletin 112

(A)

Zurich as a World City

P.J. Taylor

1. WHAT IS A WORLD CITY?

Zurich has been one of the leading international financial centres of the twentieth century. Currently doubts are being expressed as to whether it can maintain this position in the twentieth first century. Although the city's financial industries are growing faster than other smaller Swiss financial centres, Zurich in turn is growing slower than other larger international financial centres. There are two related reasons creating this situation and consequent fears for the future. First, a globalization of financial industries in which new concentration processes have benefited Zurich more within Switzerland than without. Second, the latter is exacerbated by doubts over the future geography of private banking, for long the keystone of Swiss financial prowess in the world. In this talk I wish to broaden this debate to consider Zurich as a world city.

(i) International Financial Centres, World Cities and Global Cities

Given Zurich's past financial prowess and the fact that financial services are to be found at the centre of the contemporary concept of world city we will begin with financial centres. Such centres have always been seen in plural terms: as foci for the circulation of capital, they form networks within which individual cities operate as hubs.

International financial centres emerged in the nineteenth century and blossomed in the twentieth century. Their development is premised upon the development of nation-states with associated 'national economies'. This created a mosaic economic geography wherein processes of financial concentration operated at the national level to create at least one important financial centre per country. Facilitating international investment and international trade between countries, most such centres acted as gateways, facilitating the movement of capital into and out of countries. Generally, the importance of a centre was dependent on the size of a country's economy (e.g. Paris being a more important financial centre than Stockholm) with first London and then New York dominating on the back of their respective, hugely successful, national economies. To punch above your 'national economic weight' required financial niche development of which the best example is Switzerland with its private banking specialization creating several international centres in one small economy (Zurich, Geneva, Basel).

Global financial centres transcend the mosaic of national economies. Building on the transformation of capital markets following the breakdown of the Bretton Woods agreement in 1971, nationally-based financial operations have shrunk and been replaced by globally-orientated operations (Sassen 1999). Thus 'national banks' have transmuted into 'global banks' through fusions that respect no borders. Such banks have scores, even hundreds, of branches in cities outside their home country. In this way global financial centres have become overtly transnational, the home of as many foreign banks as domestic ones. This process has accentuated centralization within countries so that what were leading domestic financial centres become even more dominant nationally due to this foreign infusion. Thus very dominant centres become completely ascendant (e.g. London over Birmingham /Manchester; Paris over Lyon) and elsewhere former important 'second financial centres' are falling behind (e.g. New York, Sydney, Toronto, Sao Paulo and Mumbai are outstripping their former rivals, respectively, Chicago, Melbourne, Montreal, Rio de Janeiro and New Delhi/Calcutta) (Sassen 1999). In Switzerland, of course, this is true of the relationship between Zurich and Geneva. The former joins with centres such as Tokyo, Hong Kong, Paris, Singapore, Chicago, Frankfurt, Madrid and Milan as a global financial centre in a stratum below New York and London who are the pre-eminent cities of this new genre.

But exactly how new is this genre. In general we can note that capital is never a respecter of boundaries (even in twentieth century war time!). Specifically global financial centres replicate many of the functions of the pre-nation-state transnational financial centres, often incorporating gateway functions also. Thus we can say that global financial centres are mimicking a geography of operations long pioneered by Zurich and other early 'global centres' (Sassen 1999). And here's the rub, with many other financial centres going global, Zurich has lost and is losing its distinctiveness, part of its successful twentieth century niche, and hence the concern for its twenty first century role. Being a financial centre, global or otherwise, may be sufficient for Luxembourg City or Hamilton Bermuda, but is it alone enough to sustain a major European city such as Zurich under conditions of contemporary globalization?

World city denotes a much broader concept than financial centre. The latter remains as a core activity but many other functions are important. Thus, Friedmann (1986), recognising that international financial centres were 'globalising' (Mayer 1981), incorporates finance in his definition of a world city hierarchy. In addition he includes other business services (e.g. advertising) but the key new element is consideration of headquarter functions of multinational corporations. With the latter creating a new international division of labour (the industrialisation of the 'Third World'), the consequent complex world economy requires control and command centres which are the core feature of Friedmann's world cities. With banks included among the multinationals, Friedmann (1995) identifies Zurich as a world city in his revised inventory (although curiously it is listed as a national articulator rather than a multinational one).

The key contribution to understanding leading contemporary cities as more than financial centres is that of Sassen (2000, 2001). Using the term global cities, she argues that leading contemporary cities are much more than co-ordinating nodes for large corporations. The geographical context changes from a new international division of labour to contemporary globalization with its enabling technologies combining communications and computers. It is in this new global space of flows (Castells 1996) that provides cities with new possibilities. They have responded by becoming sites of sophisticated production of new financial innovations and new global service products for both of which they also acts as marketplaces. The result is a dynamic new type of city (this is why the descriptor 'global' is used) that is nothing less than a global service centre for a globalising world economy. It is this concept that is built upon below in my empirical assessment of Zurich as a world city below.

Hence, we can see that world/global cities are global financial centres plus something extra, the extra critically including other business service functions. In these economic centres that are products of globalization wherein finance has a central role but where sophisticated producer services in general form the creative core of global service centres. The suggestion is that to be a successful city in contemporary globalization, leading cities cannot connect to the rest of the world just through traditional financial linkages. World cities are a much more sophisticated economic packages covering a variety of economic sectors, feeding into and off each other. There is no doubt that Zurich remains an important financial centre, the question explored here is the degree to which it has become a world city, aka a global service centre.

(ii) Cities and Clusters

Processes of contemporary globalization are premised upon the linking of information and communication technologies. This telematics eliminates distance as a factor in the transmission of information and has led to a reorganisation of economic activities wherein information is important. In particular the new infra-structural technologies have enabled a decentralization of functions. Virtual markets, not least financial markets, have been said to denote the end of geography. No need for cities then .. and yet globalization in practice has been associated with a spectacular rise in the economic importance of many cities across the world, world cities no less. Why is it that cities rather than computers remain the key media through which business and finance are co-ordinated and strategic deals brokered (Sassen 1999)?

The answer to this conundrum has to do with the nature of cities as clusters of business and financial activities (Sassen 2000). Cities are not just places where business and financial transactions take place, concentration of such activities itself provides added value to business. Cities constitute knowledge-rich environments that give those businesses locating there an advantage over outsiders. Thus cities are innovation centres, forever changing as dynamic learning mileaux for leading economic sectors. It is this role that has been enhanced by globalization. Thus rather than the fluidity of information harming cities, their contemporary role has been enhanced with the necessary information/knowledge becoming more and more complex as markets, and influences on markets, grow in size and reach. The clustering of different economic activities is by no means new, but it has taken on a new significance with globalization, so much so that this era has been 'the age of cities' (Martin and Schumann 1997).

The key point, of course, is that not all cities have benefited equally from this new concentration of high value-added economic activity. It is generally agreed that the two chief beneficiaries have been London and New York. Other cities that have done well are Tokyo, Hong Kong and Singapore although the Pacific Asian financial crisis of the late 1990s reminds us that the dynamism in the network can work both ways, Friedmann (1995), in particular, has emphasized the volatility in winners and losers among world cities. But on the other hand, it is the nature of clusters that they cannot be easily replicated elsewhere. They define special places with embedded historico-cultural advantages for the conduct of business. Clusters cannot be easily swept away but equally clusters are not eternal.

(iii) Cities and Flows

City clusters are only half the story of successful world cities. Cities have always been the meeting places of their wider society. Historically, cities operate in spaces of flows organized as networks: it would seem that a lone city (i.e with no flows to and from other cities) is an impossibility. But is this still true? With the fluidity of information transmission why doesn't the servicing of global capital operate through one great city encompassing the mega-clusters of all necessary sectoral expertises? With a mega-knowledge-rich environment generating a surfeit of innovations in an unprecedented dynamic place, this one great world city could transmit its great thoughts and ideas across the world. Other cities would then just have the role of receivers; operationalizing the messages from afar. In fact this could hardly be further from the reality of spatial organization under conditions of contemporary globalization. We are living in an age of cities in the plural not the singular. Why the continued plurality of world cities that operate, as I will show below, as a global network of strategic places?

The answer to this conundrum again rests upon the need for knowledge, this time relating to the geographical complexity of the world economy. Although all world cities are cosmopolitan, they each have unique strengths relating to markets within their local, national and regional reaches. Since world markets are constituted as congeries of markets at these smaller geographical scales, it is impossible to efficiently centralize economic activities. For instance, Hong Kong is the place to be for businesses wanting to take advantage of the massive emerging market that is China. Knowledge of opportunities and pitfalls within the changing dynamics of doing business in China are honed in Hong Kong not London or New York. This process operates around the world and has led to the rise of important world cities in all emerging markets: Mexico City, Sao Paulo, Mumbai and Moscow are among the most important.

Thus, once again, the relevance of cities, this time as strategic places, has been advanced through processes of globalization. In this case the success or failure of cities relies on the flows coming in and out of the city, the densities of which depends on both the vitality of a city's economic clusters and the growth of its encompassing markets. Flows and clusters are therefore indelibly linked in the development of cities. The two most important cities - London and New York - owe their contemporary success both to outstanding, vibrant economic clusters and to being the leading crossroads in the global space of flows.

2. LESSONS FROM LONDON?

The Globalization and World Cities (GaWC) Study Group and Network (centred at Loughborough University, UK) has carried out three projects on London in the last few years that all may have relevance to how we think about the future of Zurich. The first project was about how London and New York coped with the Asian financial crisis of the late 1990s. The main finding was that in a globalising economic world, London and New York were not 'outside' the action, they were integral to the unfolding of the crisis through the world city network (Beaverstock et al. 2002). The second project involved seeing how Frankfurt fared in relation to London with the launch of the Euro and the location of the European Central Bank in Frankfurt. The main finding was that Frankfurt and London were not really in competition, their relationship was one of complementarity so that what was good for London was good for Frankfurt and vice versa (Beaverstock et al. (2001). The third project investigated the economic clusters in central London. The main finding was that vibrant economic clusters were various and vital to London's pre-eminence as a global city (Taylor et al. 2003). I will elaborate on these findings by emphasising three main points.

(i) The Whole More than the Sum of the Parts

This is a crucial feature that appears again and again in our research. London is constituted as a large critical mass in terms of both clusters and flows. This was particularly apparent in the comparisons with Frankfurt. To think that the launch of the euro should harm London because the UK stayed outside Euroland is to totally misunderstand London's position in the world city network. Frankfurt is simply not a rival to London in any overall assessment. Certainly, over several decades, the German city has grown in importance with the German economy but this has no where near placed it in a position to replace London as Europe's global city. In the research many of our respondents totally derided this idea and denigrated Frankfurt's global pretensions: it operates at a completely different scale. The situation was more a division of labour with London at the centre of global flows and Frankfurt crucial for central European business.

This size problem is common to many of Europe's world cities. In global terms they are only medium-sized metropolises which means they cannot sustain the full range of services that London provides in the world economy. As well as its leading role in finance, London has critical clusters of business in a range of other advanced services, such as advertising and law, plus critical media and tourism and other world class functions. Simply put, it has to be recognised that for the foreseeable future London will be Europe's global city and other European cities will have to adapt their world role to this situation.

Zurich is an archetypal European medium-sized city providing global services. It cannot compete with London (or New York, Tokyo, etc.) as a global city but it can find a niche to operate in the world city network alongside global cities.

(ii) Strategic Collaboration

This means that we have to rethink the simple idea of city competition. There is no zero-sum game that cities play against one another. The relationships between competition and co-operation are complex and highly variable over time and space. The one constant is the perennial under-estimating of the importance of co-operation. This is largely because we have become accustomed to viewing cities as 'national hierarchies' (i.e. competitively) whereas seeing them within a world city network implies a different set of inter-city relations that emphasize collaboration.

Sassen (1999) gives a very good example of this process. Her trilogy of global cities are complementary is two distinctive ways. As well as their well-known complementary geographical locations for enabling the '24-hour financial market' to exist, there is also strategic collaboration at the full global level. According to Sassen London provides the global platform through which financial practices are harnessed, New York is the source of key financial market innovations - 'the Silicon Valley of finance', no less - and Tokyo provides the money as a sort of 'raw material' from a 'plantation economy'. In other words, even Sassen's leading global cities do not attempt to have 'Jack of all trades' roles. As nodes in a network, every city is different, successful global cities do have a wide range of roles but their world-class clusters are each specialised and unique. For instance, in resolving the Asian crisis it was New York banks and London law firms that were strategically important.

Zurich needs to address what its collaborations are in the world city network in order to maintain its position as a successful strategic place.

(iii) "Wimbledonization"

This is a term used to describe London's financial prowess through analogy with the world's number 1 tennis tournament. Just as in the tennis where British players seldom do well, so with the City of London, British banks are not the reason why the City is so successful. With the exception of HSBC, banks with London headquarters are not leading global players and yet London is a leading global financial centre. In other words, like the tennis tournament, it is the place to be - all financial firms with global pretensions have to be located there. Not long ago there were rumours that Deutche Bank was thinking of moving its HQ from Frankfurt to London. This caused great consternation in Frankfurt but was largely irrelevant to London. What was important to the latter was the fact the Deutche Bank's global operations took place in London. Where the bank's HQ was location was beside the point.

Wimbledonization is a further indication that we should not think in terms of crude inter-city competition. The story of the Liffe takeover illustrates this well. Iniitially the London Stock Exchange planned to buy Liffe in order to consolidate London financial markets. Before the deal could be completed their Frankfurt rivals emerged as potential rivals. In the event, however, Liffe was bought by Euronext (the Paris-based combination of French, Dutch and Belgium exchanges). To consider this as a victory for Paris over London would be a misleading judgement because (as opposed to the LSE) London did well out of the transaction. On completion of the deal, Euronext decided to relocate all its derivative trading to Liffe in London thus exhancing the latter's global role in derivative exchanges. Quite simple, irrespective of its Paris HQ, Euronext decided that London was the place to be for this particular part of its operations (Leader 2001).

Headquarter functions can be important but thinking of cities in terms of the world city network means that it is niches that are more important: what sort of 'strategic place' can Zurich become?

(iv) Not a world of "mini-Londons"

Finally, the question mark at the end of this section's heading needs to be reinforced. Transferring 'lessons' from one city to another in a network can only be meaningful at a fairly general level. My emphasis on clusters and flows is intended to be at this level - whaichclusters and flows are appropriate will vary from city to city. This is not an area where the old planning ideal of diffusing best practice will work: emulation cannot create specific roles in networks. As Krätke (2000) has shown for post-1989 Berlin, rushing to become a new 'service metropolis' like London is a recipe for disaster resulting in gross over-provision of offices.

In the world city network, selective positionality is all - for Berlin, and for Zurich.

3. THE WORLD CITY NETWORK

A fourth GaWC project has attempted to specify and measure the world city network. I will report results concerning Zurich from this quantitative project but first I will briefly outline the basic methodology so that the findings will make some sense.

(i) An Interlocking Network

What sort of network do world cities constitute? Most networks are quite simple and consist of just two levels: the nodes and the set of interconnections that is the network. With the world city network however there are three levels. Clearly the cities define a nodal level and their interconnections form a network level at the scale of the world economy but in addition there is a critical sub-nodal level. This is defined by the global service firms (providing financial and business services) that cluster in the cities and whose global practices link cities together. Here I am thinking of the myriad flows between the office blocks that typify world city skylines: flows of information, knowledge, ideas, plans, strategies, instruction, and personnel. In order to provide a seamless service that protects their global brand name, these large service firms have office networks across large numbers of cities. In this way the service firms 'interlock' cities through their intra-firm connections. Thus the world cities constitute an interlocking network in which the sub-nodal level, the firms, are the critical agents in network formation. For more details of this network specification see Taylor (2001a).

Once specified, we can begin the task of measuring the network. This requires information on the global strategies of global service firms as reflected in their city office networks. Defining global firms as those having offices in at least 15 different cities with at least one in each of northern America, Pacific Asia and western Europe, we have collected such data on 100 global service firms (18 in accountancy, 15 in advertising, 23 in banking/finance, 11 in insurance, 16 in law and 17 in management consultancy). For each firms we codified their presence in 315 cities on a scale from 0 (no presence) to 5 (very important presence, their headquarters). We term these numbers service values since they indicate the importance of a city to a firm's global operations. The result is a 100 x 315 service value matrix. Each column in the data matrix defined a firm's global strategy as an array of their 315 service values and each row describes a city's service mix as an array its 100 service values. The data were collected in the summer of 2000, for further details see Taylor et al. (2002a).

(ii) Network Connectivity

Once in data matrix form, the world city network can be measured and analysed. The most basic measure in network analysis is connectivity between the nodes. In this interlocking network, connectivities are calculated as sums of products through comparing the service mixes of cities. The first underlying assumption is there will be more flows generated between two cities that house the same firms than between two cities that share few common firms. The second assumption is that the more important offices generate more flows and that for inter-city flows this is a multiplicative relation. By aggregating a city's connectivity with all other 314 cities, we produce a measure of the city's global network connectivity. For an algebraic derivation of this measure see Taylor (2001a).

We are now ready to present some results. Table 1 shows the top 30 world cities in terms of global network connectivity. The dominance of London and New York is not surprising but Tokyo's fifth position is unexpected given Sassen's (2001) identification of this city as one of the global city trilogy. Generally speaking, however, the ordering is much as expected but the key point is that before this research there has not been a ranking of world cities on such a comprehensive basis. Thus Zurich's position at 19 with a little less than one half of London's connectivity is a new result that can stand as a starting point for discussion of the city's twenty first century future.

4. ZURICH'S GLOBAL NETWORK CONNECTIVITY

As a measure computed by aggregation, network connectivity can be disaggregated into constituent parts. This enables us to understand the make up of any city's connectivity - where it is strong and where it is weak. This is what I have done for Zurich along with parallel examinations of Frankfurt and Geneva for comparison. These two cities are chosen for straightforward reasons - Frankfurt is a medium-sized European city that has proved to be more successful than Zurich in the world city network; Geneva is Switzerland's other world city.

(i) Disaggregating Global Network Connectivity

As we have seen the global service firms belong to six different service sectors. This is the obvious first way in which we can disaggregate Zurich's overall connectivity. Table 2 shows the six sectors along side the global connectivity for our three selected cities. There are both expected and surprising results in this table. First, it can be noted that Zurich ranks above Geneva in all sectors. The closest Geneva gets to Zurich's rank is for law and this is Zurich's weakest sector. Notice that the city's strongest sectors are insurance and management consultancy with banking/finance not as strong as might be expected. I think this is because the data focuses on global service firms and although Swiss major banks qualify many other's do not. However the further implication is that non-Swiss global banks do not generally treat Zurich as a major node in their networks. This is in contrast with Frankfurt where its world ranking improves appreciably when just taking banking/finance into account. A further indication of Frankfurt's success (and contrast with Zurich) is its very high ranking for global legal services - the latter are particular distributed in global financial centres. In conclusion, as a medium-sized world city Zurich's only discernible global service niche in this data is in insurance.

There are other more subtle ways in which we can disaggregate connectivities. Table 3 shows two examples of this that attempt to measure the power that is exercised through a city: nodes that act as network hubs (for a detailed desciption of this measures see Taylor et al. (2002c). Command is measured by just computing connections for firms headquartered in a city (i.e. with service values of 5). The results define command centres of which there are only 21 such cities in the data. Table 3 shows that Geneva is not one of these (i.e. Geneva houses none of the headquarters of the 100 global firms). It also shows that Zurich ranks high on this measure, ranking only just behind Frankfurt. This reflects the leading Swiss banks and insurance companies included in the 100 global service firms. However, when we consider a more general measure of power, dominance, a rather different picture appears. In this measure all connections are included where a given city has the higher service value in a dyad (it will, by definition, include all connections that were used to measure command). Table 3 shows that for this measure Zurich's ranking tumbles whereas Frankfurt's stays the same. This indicates that beyond Swiss global service firms, Zurich does not figure prominently in firms' global strategies whereas Frankfurt does. In conclusion, Zurich's connectivity is over-dependent on home-based global service firms.

Finally, we can look at the connectivity that is derived from subordinate connections. This is where the city in a dyad has the lower service value. High values of subordinate connectivity indicate a city where global service firms do not locate important offices but nevertheless do feel they have to have an office in the city. To understand this measure, the top ten ranked cities are included as well as our selected cities in Table 4. The process operating to generate these 'places to be' appears quite straightforward: 8 cities in the top ten are gateway cities to 'emerging markets' epitomised by the top pair Beijing and Moscow. Brussels appearance at 8^th is explicable in terms of its particular political role in the EU as the world's largest bounded market. The really big surprise, therefore, is Zurich's third place on this list. Note that in Table 4 Zurich far outstrips Frankfurt on this measure as does Geneva. One possible explanation is Zurich's role as a centre of the Swiss tradition of private banking with their links to the Third World. Although these banks do not qualify by our criteria as global service firms, their clustering in Switzerland does attract many of our global service firms. In this interpretation, Zurich appears as a global hub for emerging/Third World markets.

(ii) Where in the World: Zurich's Hinterworld

We are familiar with the idea that every city has its own local hinterland. However in addition, cities today have connections that are worldwide. To capture this idea I have coined the concept of a city's hinterworld (Taylor 2001b). This is based on a disaggregation of global network connectivity back to its fundamental constituent parts, the dyads - how a given city relates to every other city.

Table 5 shows the top 30 dyad city connections to Zurich. As we would expect this list is very similar to the global connectivity list (Table 1): London and New York have the largest connectivities to Zurich. But the list is not exactly the same: London and New York's ranking are reversed for example. Table 5 shows all the differences in ranking indicating several positive changes for Pacific Asian cities. This reflects the fact that the cities of the Pacific Asian region are particularly connected into the world city network through the banking/finance sector (Taylor et al. 2002b). However comparing differences in rankings is not the best way of finding the particular geographical strengths and weaknesses of a city's hinterworld.

An alternative approach 'removes' the global network connectivities to reveal the particularities in the pattern of a city's connectivity to other cities. Such measures of relative hinterworld can be produced by regressing the absolute hinterworld values (as in Table 5) against global network connectivity scores (as in Table 1) and computing the residuals. This indicates where a city is over-linked (positive residual) with respect to global network connectivities and where it is under-linked (negative residual). For this exercise we focus upon only on the top 123 world cities (they all have connectivity at least one fifth of London's) and look at the connectivity of each of our 3 selected cities across the world. For further details of this approach see Taylor and Walker (2003).

Zurich's hinterworld is shown in Figure 1. In general, the first point to make is that Zurich most certainly does not have a local (i.e. European) bias to its connectivities: it is relatively over-linked to cities across all areas of the world. The only region where the cities are all relatively under-linked is western USA. The cities of Pacific Asia are well represented reproducing the evidence in Table 5. In specific terms, New York and especially Chicago are over-linked while Paris and especially London are under-linked. Zurich's relatively strong links with Germany are to Berlin and Munich in contrast to under-linkage to Frankfurt and Hamburg. But to fully appreciate the nature of this hinterland it is necessary to make comparisons. Frankfurt's hinterworld is shown in Figure 2. Three features stand out. First, like Zurich, there is a bias towards Pacific Asian cities for over-linkage but it is even stronger here. Second, there is more of a concentration in Europe, specifically central Europe (Frankfurt is strongly over-linked to all other German world cities) and eastern Europe. Third, there is little or no important over-linkage to Third World cities (Rio de Janeiro being the only exception). In contrast, Geneva's hinterworld (Figure 3) shows a strong southern European bias with both northern America and Pacific Asia cities largely missing from the over-linked category. This is compensated for by much more over-linkage in Third World cities. These include banking centres - as well as Nassau and Manama where Zurich is also over-linked, Port Louis is featured. There are strong linkages to Latin American cities plus a scattering of major cities across the Third World: Cairo, Abu Dhabi, Istanbul and Mumbai. Given that both Swiss cities have unusually large numbers of over-linked cities in the Third World, we can surmise that this is related to the country's traditional role as a global centre for private banking. However, overall, these comparisons lead to the conclusion that Zurich's hinterworld is a sort of compromise between Frankfurt's basic First World financial centre pattern and Geneva's quite unusual bias towards Third World cities.

5. CONCLUDING COMMENTS

What does this all mean for Zurich in a world city network formed under conditions of contemporary globalization? At this time I do not want to repeat the detailed findings and derive implications. The key message of the paper can be summarised in two points, one general and one specific.

The general point is that we cannot consider Zurich, or indeed any other city, on its own. Metropolitan policymaking is inevitably place-based given the nature of the local politics in which it is grounded but the fact remains that it must keep in mind the network dimension, In any scenario building, the rest of the network will not stop still while Zurich implements place-based policies to make itself more attractive. If all its rivals do the same, then the city will be working hard to stay where it is. So-called rivals are better viewed as complementary nodes in a world-wide space of flows who can, as well as taking business off Zurich, be instrumental in steering business Zurich's way. The latter is clearly a preferable process for all concerned but is difficult to 'plan' using the usual municipal instruments.

The specific point is that to be complementary to other nodes in the network, Zurich has to cultivate clusters and flows that provide it with specific niches. This means that business will 'naturally' flow to Zurich from other world cities for is specialism while business will be passed on from Zurich to other cities where they have the appropriate expertise. The empirical evidence presented above does show Zurich to be unusual as a European world city and this is born out in other more general analyses (for instance, in a simple discriminant analysis Zurich is misclassified as 'western European' and is reallocated to the eastern European category - see Taylor (2003)). And this specificity within Europe has to be seen as an asset, something to build upon as the world city network intensifies through the twenty first century.

References

Beaverstock, J V, Hoyler, M, Pain, K and Taylor P J (2001) Comparing London and Frankfurt as World Cities. London: Anglo-German Foundation

Beaverstock, J V, Doel, M A, Hubbard, P J and Taylor, P J (2002) 'Attending to the world: competition, cooperation and connectivity in the world city network' Global Networks 2, 111-32

Castells, M (1996) The Rise of Network Society. Oxford: Blackwell

Friedmann, J (1986) 'The world city hypothesis' Development and Change 17, 69-83

Friedmann, J (1995) ' Where we stand: a decade of world city research' in P L Knox and P J Taylor (eds) World Cities in a World-System. Cambridge: Cambridge University Press

Krätke, S (2000) 'Berlin: the metropolis as a production space' European Planning Studies 8, 7-21

Martin, H-P and Schumann, H (1997) The Global Trap. London: Zed

Sassen , S (1999) 'Global financial centres' Foreign Affairs 78, 75-87

Sassen , S (2000) Cities in a World Economy. Thousand Oaks, CA: Pine Forge

Sassen , S (2001) The Global City. Princeton NJ: Princeton University Press

Taylor, P J (2001a) 'Specification of the world city network' Geographical Analysis 33, 181-94

Taylor, P J (2001b) 'Urban hinterworlds' Geography 86, 51-60

Taylor, P J (2003) 'Regionality in the world city network' International Sociologist

Taylor, P J, Catalano, G, and Walker, D R F (2002a) 'Measurement of the world city network' Urban Studies 39, 2367-76

Taylor, P J, Catalano, G, and Walker, D R F (2002b) 'Exploratory analysis of the world city network' Urban Studies 39, 2377-94

Taylor, P J, Catalano, G, Walker, D R F and Hoyler, M (2002c) 'Diversity and power in the world city network' Cities 19, 231-41

Taylor, P J, Beaverstock, J V, Cook, G, Pandit, N and Pain, K (2003) Financial Services Clustering and its Significance for London. London: City of London Corporation

Taylor, P J and Walker D R F (2003) 'Hinterworlds revisited' Geography 88

Table 1: Global Network Connectivity: Top 30 Cities

Table 2: Connectivities by Sector: Zurich compared with Frankfurt and Geneva

Table 3: World City Hubs: Zurich compared to Frankfurt and Geneva

Table 4: "The Places To Be": Subordinate Connectivity

Table 5: The 30 Cities Most Connected to Zurich

Figure 1: Zurich's Hinterworld 

This cartogram places cities in their approximate relative geographical positions. The codes for cities are:

AB Abu Dubai; AD Adelaide; AK Auckland; AM Amsterdam; AS Athens; AT Atlanta; AN Antwerp; BA Buenos Aires; BB Brisbane; BC Barcelona; BD Budapest; BG Bogota; BJ Beijing; BK Bangkok; BL Berlin; BM Birmingham; BN Bangalore; BR Brussels; BS Boston; BT Beirut; BU Bucharest; BV Bratislava; CA Cairo; CC Calcutta; CG Calgary; CH Chicago; CL Charlotte; CN Chennai; CO Cologne; CP Copenhagen; CR Caracas; CS Casablanca; CT Cape Town; CV Cleveland; DA Dallas; DB Dublin; DS Dusseldorf; DT Detroit; DU Dubai; DV Denver; FR Frankfurt; GN Geneva; GZ Guangzhou; HB Hamburg; HC Ho Chi Minh City; HK Hong Kong; HL Helsinki; HM Hamilton(Bermuda); HS Houston; IN Indianapolis; IS Istanbul; JB Johannesburg; JD Jeddah; JK Jakarta; KC Kansas City; KL Kuala Lumpur; KR Karachi; KU Kuwait; KV Kiev; LA Los Angeles; LB Lisbon; LG Lagos; LM Lima; LN London; LX Luxembourg; LY Lyons; MB Mumbai; MC Manchester; MD Madrid; ME Melbourne; MI Miami; ML Milan; MM Manama; MN Manila; MP Minneapolis; MS Moscow; MT Montreal; MU Munich; MV Montevideo; MX Mexico City; NC Nicosia; ND New Delhi; NR Nairobi; NS Nassau; NY New York; OS Oslo; PA Paris; PB Pittsburgh; PD Portland; PE Perth; PH Philadelphia; PL Port Louis; PN Panama City; PR Prague; QU Quito; RJ Rio de Janeiro; RM Rome; RT Rotterdam; RY Riyadh; SA Santiago; SD San Diego; SE Seattle; SF San Francisco; SG Singapore; SH Shanghai; SK Stockholm; SL St Louis; SO Sofia; SP Sao Paulo; ST Stuttgart; SU Seoul; SY Sydney; TA Tel Aviv; TP Taipei; TR Toronto; TY Tokyo; VI Vienna; VN Vancouver; WC Washington DC; WL Wellington; WS Warsaw; ZG Zagreb; ZU Zurich

Figure 2: Frankfurt's Hinterworld 

For city codes see Figure 1

Figure 3: Geneva's Hinterworld 

For city codes see Figure 1