Information transmission in international finance certainly did not start with Nathan Rothschild, but when he was the first in London to hear the news of Napoleon's defeat at Waterloo, allegedly making a fortune of it, the episode soon became one of the industry's favourite legends grounded in information technology. In the early 19th century, in the financial industry information technology was working basically according to the same principles we know today such as digital codes, data compression, error recovery and encryption. The talking is not about computers and the internet, but of "the mother of all networks" as one author puts it, the optical telegraph. The optical telegraph was followed by the electric telegraph, the telephone, the fax machine and, eventually, the computer. As the following chapter shows, each time, financial institutions all around the world eagerly jumped at the opportunity to use the new medium for speeding up communication and trade. The recent mania about electronic exchanges and e-commerce firms offering financial information and the opportunity to buy and sell stocks, bonds, derivatives and other financial instruments online to a broad public sometimes makes forget that this has been the usual way of doing business for large parts of the financial industry for many years.
The emergence of new information technologies and "virtualisation" of activities is often said to reduce the importance of location for the financial industry and diminish the role of world cities such as London, Paris and Frankfurt as financial centres. This would have severe consequences for economic growth. At the beginning of the 21st century cities are generally struggling to cope with the effects of two phenomena, the transition from an industrial and manufacturing-based economy to one in which services and information and communication technologies play a dominant role and the competitive pressures of increasing globalisation. In this situation, financial services are one of those few sectors promising continuing growth, employment and tax incomes to their communities. A shrinking importance of location would threaten to reduce these benefits.
In the literature, the view that virtualisation and new information technologies reduce the role of cities and lessen discrepancies between 'cores' and 'peripheries' in the world economy is widespread (Graham and Marvin, 1996; Moss and Townsend, 1998). But, mapping internet traffic and activity shows the surprising result that, in contrast to this belief, hierarchies appearingly become even more pronounced. This holds in particular for financial centres. Manhattan has some of the highest densities of domain names in the world, Singapore and Hong Kong have become critical nodes in an Asian-Pacific financial network geography, and in Europe, after far-reaching deregulation in the 1980s, London developed into a centre for all kinds of information-rich financial services. In the literature, explanations for this phenomenon stress the role of headquarter and control functions, of a reliance on producer services, and of cities providing a platform for global operations (Daniels 1993, Sassen 1999). While some financial services are decentralising with rising information and communication technologies, the overall need for centralisation of control and coordination in the sector is rather reinforced further enhancing financial agglomeration and concentration. But, the relations between information technology, financial services and city growth are more complex as the study of concepts of the nature of virtual reality, and the linkages between real and virtual processes, shows. In order to explore the implications for the world's financial landscapes an answer has to be found to the following questions: What characterises virtual activities in general and in which way do they differ from traditional ones? What is it that gets lost, and what is won, in the process of virtualisation of financial services and functions? To what extent can electronic flows replace physical activities, and to what extent and in which ways do they complement them, in the world of international finance?
The paper will proceed as follows. Section 2 will present a short chronology of information technologies in the world of finance from ancient to modern times. Since most of the discussion on the relation between finance and IT is limited to the aspects of communication and networking, section 3 will show in detail the implications of the recent convergence of information transmission and processing technologies covering a wide range of areas and activities both in traditional financial markets and new market segments and structures. It will describe developments in wholesale and retail banking and securities trading, trace the industry's process from internationalisation to globalisation and study the implications of the virtual dimension added by electronic linkages and the internet focusing on activities such as customer services, trading, risk management and settlement. Then, section 4 will look at the nature of virtual reality, and the relation between real and virtual spaces and places in general, and analyse how these are altering our view of the world's financial landscapes of centres and peripheries. Section 5 will draw some preliminary conclusions about the lessons for the future prospects of the growth of cities in the age of electronic finance.
IT IN THE WORLD OF INTERNATIONAL FINANCE - A SHORT CHRONOLOGY
The use of information technology to collect, generate and record financial data is nearly as old as the technology itself, spanning thousands of years from the earliest forms of pictorial representation and writing to the latest advances of electronic storage and transmission. From the early beginnings, financial activities were not restricted to particular locations but extended over regions and continents, and great efforts were made to overcome the limits of time and space in long-distance communication in order to transmit news about latest political and economic developments, natural catastrophes and other unforeseen events affecting markets and prices and signalling business opportunities.
Prior to the invention of telegraphy in the 19th century, information was bound to move at the same speed, and over the same distance, as the prevailing transport system would allow (Dicken, 1998, p. 151). Financial activities were largely determined by personal knowledge of people and circumstances (Favier, 1992, pp. 24 ff.). As one author put it:
Success in money and banking operating in a number of countries ... required having a large number of brothers or cousins, with a single combined interest and thinking more or less alike, to solve the agency problem. (Kindleberger, 1993, p. 258)
The most common method to communicate over long distances was to hire a person to deliver a message as fast as possible, either a human runner or a rider on horseback. Safety considerations made early rulers place guards at regular distances along the roads. They became the forerunners of the relay systems. References to messenger systems were found dating back almost 4000 years to ancient Egypt and Babylon. The Romans had relay stations which kept a reserve of 40 horses and riders. And for Asia, Marco Polo reported of a system of post-horses used by the Mongol ruler Kublai Khan (1215-1294) with about 200 horses per post which reached considerable speed - in this it was compared to the Pony Express which operated much later (from April 1860 to October 1861) in the United States covering the 3,200 km distance from Missouri to California in about ten days (Holzmann and Pehrson, 1994). Other means of transmission in history were homing pigeons, mirrors, flags, fire beacons and semaphores. For example, it was a pigeon that brought Nathan Rothschild the news of Napoleon's defeat. Mail was delivered by stage coach, caravans and merchant vessels. Travellers were routinely ask to take messages with them. When young Pierpont Morgan left England to go to school in Vevey, Switzerland, travelling over Calais and Paris in 1854, he was asked by the American minister in London, James Buchanan, to deliver a packet of government papers to Paris which was not unusual. (Strouse, 1999, p. 54)
Telegraph and Ticker Tape
The arrival of the telegraph made all the difference allowing messages to be sent with great speed over very large distances. The first optical telegraph line started operating in France between Paris and Lille in May 1794. Soon other European countries followed and in 1830 "lines of telegraph towers stretched across much of western Europe, forming a sort of mechanical Internet of whirling arms and blinking shutters" (Standage, 2000, p. 18). But the system had also its drawbacks. It was expensive to run requiring shifts of skilled operators at each station and involving to build towers all over the place. Beside, optical telegraphs would not work in the dark or in fog and mist.
Eventually, it were electric inventions that changed the world. The discovery of the electric telegraph in the 1830s, and the telephone in the 1870s, marked a distinct new era. Discovered simultaneously in Great Britain and the United States, the telegraph made possible the first efficient direct control of operations in one distant place from another. It became the first agent of instant communication between countries and continents coordinating international financial and commodities markets (Mokyr, 1990, p. 124). For example, prior to its introduction between New York and Philadelphia, the transmission of price data from the distant market to the local one had taken one day. The telegraph enabled investors to obtain the price information before their own market closed. Before telegraph lines were established between New York and New Orleans, dealers in foreign exchange in one city received quotes from the other with a delay from four days to a week. This was reduced by the telegraph to a day or less (Garbade and Silber, 1994). The telegraph fundamentally changed America's financial landscape. In 1850, there were 250 stock exchanges. 50 years later, New York had become the dominant exchange standing out as the only financial centre of national importance (Edwards 1998).
However, it was not until after 1848 that the impact of the new technology on financial markets was really visible. Its expansion was fastest in the United States. In New York, in these years, there were eleven separate lines and "it was not uncommon for some bankers to send or receive six or ten messages each day." (Standage, 2000, p. 58) In contrast, in Britain the telegraph was at first more associated with the railways. Nevertheless, when London reported the first congestion problems in the early 1850s, half of all telegraph messages were already related to the Stock Exchange. Soon, there were telegraphic money transfers. However, one big problem was security. Banks had their own sophisticated private codes for money transmission but, up to the implementation of a scheme developed by the Western Union (the then dominant US telegraph company) in 1872, transferring money was considered as highly unsecure demanding a high level of trust between both parties and telegraph operators.
Financial market participants have been among the early users of telegraph facilities, but to them the real watershed became the submarine cable. London became linked to Paris in 1851, to New York in 1866 and Melbourne in 1872 (Inwood, 1998, p. 480). The transatlantic cable was used to arbitrage the London and New York securities markets immediately after its opening. Some days after the event the New York Evening Post started publishing price quotations from the London market (Garbade and Silber, 1994). In those years, English investors held a substantial volume of US Treasury debt which traded in London as well as New York. Before the establishment of the submarine link, information travelled with a time delay equal to the duration of an ocean crossing, or about three weeks. And, since purchase and sale orders directed to the foreign market had to cross the Atlantic, too, execution took the same time once again. After the opening of the transatlantic cable those delays were reduced to one day. By the 1890s, telegrams between the London and the New York Stock Exchanges took three minutes from sender to receiver (Headrick, 1988, p. 104).
Britain became the leader in the world submarine cable business. British firms laid most British and non-British cables in the world and owned 24 of the world's 30 cable ships which earned the country an overwhelming military and economic advantage. London became not only the centre for financing most of the international submarine cable business but also the major communications node reinforcing Britain's position as the foremost naval, commercial and financial power in the world. For example, French journalists must learn that "news of commercial importance - commodity prices, contracts, ships' arrivals and departures, etc. - passed through London before reaching Paris. British newspapers and the Reuters news agency received reports of world conditions sooner and in more detail than their French counterparts." (Headrick, 1988, p. 114)
The telegraph changed the newspaper business. Prior to its invention, newspapers mostly covered a small locality with news of distant places taking often weeks to reach the readers. Only few larger newspapers had correspondents in foreign countries, and their letters took weeks to arrive as well. With the telegraph news from distant places became available instantly, and the question arose of who ought to be doing the reporting (Standage, 2000, pp. 140 ff.). For newspapers one obvious solution was to form groups and cooperate, establishing networks of reporters. The advantage was a far greater reach without the expenses of maintaining own reporters in distant places. In the United States, the first news agency, the New York Associated Press, a syndicate of New York newspapers, was set up in 1848. In Europe, there had already been a correspondents network even before the arrival of the telegraph. Established by Paul Julius Reuter it used carrier pigeons to supply business information and prices of bonds, stocks and shares. Initially, it operated between Aix-la-Chapelle and Brussels. When England and France were linked by telegraph, Reuter "followed the cable" moving to London as the financial capital of the world and the centre of the telegraphic network.
Communication facilities provided by the telegraph were often incomplete. For example, when telegraph communication between New York and Philadelphia - where securities were dually listed on both stock exchanges - began, the line first reached only from Newark, New Jersey to Philadelphia. Then, within a few months, the line was extended to Jersey City with messages ferried across the Hudson river to Wall Street (Garbade and Silber, 1994). In Europe, there was a similar picture. For instance, Paul Julius Reuter supplemented incomplete telegraphic communications systems on the continent by using carrier pigeons to bridge the gaps (Hall and Preston, 1988, p. 45).
In those years, telegraph companies started offering regular bulletins containing a digest of the morning papers, or a summary of the most recent market prices to the business community. But, in contrast to other businesses, for the financial industry daily or twice-daily reports soon proved not enough. Demand for more frequently updated information led to the development of the stock ticker, a machine producing a continuous record of stock price fluctuations printed on a paper tape. The name of the machine was derived from the characteristic chattering sound it made. The ticker was a great success which immedeately found hundreds of subscribers throughout New York's financial district.
These days, the telegraph is often compared to the internet:
During Queen Victoria's reign, a new communications technology was developed that allowed people to communicate almost instantly across great distances, in effect shrinking the world faster and further than ever before. A worldwide communications network whose cables spanned continents and oceans, it revolutionised business practice, gave rise to new forms of crime, and inundated its users with a deluge of information. Romances blossomed over the wires. Secret codes were devised by some users, and cracked by others. The benefits of the network were relentlessly hyped by its advocates, and dismissed by the sceptics. Governments and regulators tried and failed to control the new medium. Attitudes to everything from newsgathering to diplomacy had to be completely rethought. Meanwhile, out on the wires, a technological subculture with its own customs and vocabulary was establishing itself. (Standage, 2000, p. 1)
And, the first intranets emerged as well: From the 1870s onwards big companies with several offices began leasing private lines for internal communication. The advantage was that in this way internal messages could be sent for free allowing organisations to be controlled from a head office. This laid the foundations of today's large hierarchical companies and financial conglomerates.
Telephone and Telex
With the invention of the telephone in the late 1870s, the telegraph increasingly lost out to the new technology although it remained the main form of long-distance international communication until well into the 20th century (Hall and Preston, 1988, p. 44). Initially, the telephone was widely seen as another, improved kind of telegraph - a "speaking" telegraph -not a wholly distinct technology (Standage, 2000, p. 185). Again, business use started in the United States where by 1887, only ten years after commercial introduction, "there were 743 main and 444 branch exchanges connecting over 150 000 subscribers with about 146 000 miles of wire" (Hall and Preston, 1988, p. 50). In Europe, mainly due to political reasons, progress was much slower. In 1885, for example, the United Kingdom only had about 10,000 subscribers. Bell and Edison actively engaged in the transfer of their systems to European cities. Bell himself came to Europe in 1877. Promoters of Edison's patents were closely linked to American banks in London that had the necessary financial resources, organisations and foreign contacts to move technology across national boundaries (Hall and Preston, 1988, p. 51).
The telephone facilitated all kinds of financial and foreign exchange transactions making it much easier for the financial community to take advantage of discrepancies in rates prevailing at the same time in different locations. As a consequence, those differences tended to narrow down considerably. But, the process was handicapped by the inadequacy of long-distance telephone communications. Even after the second World War, long delays for lines for trunk calls to correspondents abroad were frequent. "It was not until the 'fifties that the improvement of long-distance telephone service and the adoption of the 'telex' system made it virtually impossible for such discrepancies to continue for any length of time between markets with identical business hours. By the 'fifties it could be said with very little exaggeration that it was almost as easy to transact business with a bank in a foreign centre as with one just across the road." (Einzig, 1970, p. 239)
In the early postwar years the telex - an international system of sending written messages where the text is typed on a machine in one place and immediately printed out by a machine in another place - became the major medium of telecommunications (Dicken, 1998, p. 155) and the single most important determinant of the world structure of financial centres. This had two aspects. First, the greater the number of direct links between centres provided by local and foreign banks via telex, the greater the quality of unfiltered information, i.e. information transmitted not through relatively public networks but through direct banking links. Second, since bankers used international telex facilities to the exclusion of nearly all other communications systems, the volume of international telex activity became an indicator of a place's international influence as a financial centre (Reed, 1994).
In the 1960s and 1970s, telex and ticker tape became of decisive importance for the growth of a new kind of financial markets, external markets or so-called euromarkets. Euromarkets, unlike conventional financial markets, occupy no fixed location but from the start relied on international telecommunications networks linking financial centres throughout the world (Langdale, 1994). Their beginnings date back to the postwar period when the widespread use of the US dollar as a vehicle currency for making payments in international transactions, the easing of exchange restrictions in major European countries and the overall growth in European business after the formation of the Common Market contributed to the need for external markets for currency deposits and loans (Dufey and Giddy, 1994). But, their true take-off is often related to the 1960s, to the restraints on foreign portfolio investment in the United States (the interest equalization tax) and on US bank lending abroad.
The Beginnings of Electronic Dealing
In the 1970s, there was another information technology revolutionising financial trading which was videotext. The technique allowed recording data on magnetic tape for to be displayed on a television making possible video conferencing - the meeting of people in various places around the world by seeing and hearing each other on a screen. But, for the world of international finance, what became even more important was screen trading. A series of companies emerged providing equipment that placed the information directly on the desks of dealers thereby threatening the traditional role of the trading floor as the centre of activity. At the beginning of the 1970, the terminals of Telerate and Quotron in the United States, and Reuters, Extel and Datastream in Europe, displayed prices fed to them by banks, brokers and dealers and allowed to change the information live, or online, at the originator's request (Hamilton, 1986, pp. 41ff.). Another revolution took place in the stock markets. In response to a crisis in OTC securities dealings in the late 1960s in the United States the National Association of Securities Dealers developed NASDAQ, an electronic dealing system. Installed in 1971 NASDAQ consisted of 20,000 miles of leased telephone lines connecting dealers' terminals with a central computing system that recorded prices, deals and other information. Trading volume soon rose dramatically and by 1985 reached more than 16 billion shares, with a value of some $200 billion, making the exchange the third largest stock exchange in the world behind those of New York and Tokyo.
Another electronic system which started in 1969 was Instinet. This system, backed by Merrill Lynch and other groups, aimed at providing a low-cost trading network along the lines of foreign exchange trading for institutions buying and selling shares in bulk which later quoted not only US stocks but also foreign stocks and options on stocks and currencies from the CBOE (Chicago Board Options Exchange). The latter was a first step towards automation of derivatives trading - a market which traditionally had been considered most resistant to being removed from the exchange floors because of the large sums involved and the volume brought by "locals" or independent floor traders. The same tendency became visible when in Europe the London International Financial Futures Exchange (LIFFE) was founded in 1982. Although keeping the open-outcry system of floor trading, from the beginning LIFFE had a high degree of automation in quotation and settlement. Later, many traditional exchanges followed this trend and invested in modern technology such that everything but the final order execution became automated (OECD, 2001).
Increased automation was accompanied by a growing tendency towards globalisation. One step in this direction was the establishment of a trading link between the Singapore International Monetary Exchange (SIMEX) and the Chicago Mercantile Exchange (CME) in 1984. This was the first of several networks and systems of an increasingly globalised automated securities trading and a forerunner of Globex, the system jointly developed by the CME and Reuters which allowed to electronically match buy and sell orders from computer terminals around the world as soon as the Chicago markets were closed (Dubofsky, 1992). The first fully electronic exchange in Europe was the Swiss Options and Financial Futures Exchange (SOFFEX) founded in 1988. Through its merger with Deutsche Terminbörse (DTB) in 1998, it became EUREX that, in 2001, was Europe's biggest derivatives market measured by the number of contracts traded. These days, outside of Switzerland and Germany, EUREX has access points in Amsterdam, Chicago, New York, Helsinki, London, Madrid, Paris, Hong Kong and Tokyo.
In the 1990s, the era of the telex came to an end and it was largely displaced by the fax, a machine used to copy documents by scanning information electronically along a telephone line and to receive copies sent in this way, and, more recently, by electronic mail using the internet to send written messages and all kinds of information and data electronically from one computer to another. But, fax and e-mail were only two of those small but highly significant changes occuring in a much larger transformation of transmission channels in global communications that took place by the development of satellite technology and optical fibres. The use of satellites for commercial telecommunications dates back to the 1960s. The Early Bird, the first geostationary satellite, launched in 1965 was able to carry 240 telephone conversations or two television channels simultaneously. The Intelsat V launched in 1995 by a multi-nation consortium of 122 countries carried 22,500 two-way telephone circuits and three television channels. When privatised in 2001 Intelsat had 450 customers in 215 countries. This and other global satellite operators now are offering broadcast, video, broadband, multimedia, internet and telecommunications services. There are regional systems as well, such as Eutelsat, and even privately owned ones like those of Citibank, Merrill Lynch, Prudential Bache and others (McGahey et al., 1994, p. 131). Transmission by satellite has the advantage that costs are largely insensitive to distance. Within the satellite beam whether transmitting for five hundred or five thousand miles, to two points on earth close together or far apart, makes no difference. But, the short hightime of satellite communications is already challenged by a new technology: optical fibre cables. Those have a still far higher carrying capacity transmitting information at very high speed and with great signal strength. Again, large financial firms are installing private networks to "bypass" public ones (McGahey et al., 1994, p. 131).
CONVERGENT TECHNOLOGIES, MARKETS AND STRUCTURES
The 1970s and 1980s saw a convergence of two initially distinct technologies, of communications technology concerned with the transmission of information on the one hand and computer technology concerned with the processing of information on the other. Computers and telecommunications became integrated into a single system of information processing and exchange (Dicken, 1998, p. 151) affecting a wide range of areas and activities such as management information systems, professional data bases, integrated text and data processing, professional problem solving, transaction clearing systems and online enquiry and electronic mail. The changes for the financial industry were substantial. Together with a far-reaching libearlisation of financial markets and capital flows in many parts of the world, the new technologies allowed financial institutions the transition from internationalisation to globalisation - from the central operation and control of worldwide activities to the dispersion of central functions to all major nodes of the world economy and their constant interaction within large networks - and, at the same time, revolutionised not only the way in which financial instruments are traded but a wide spectrum of activities from information gathering, price discovery and trading over portfolio and risk management to clearing and settlement and mergers and acquisitions.
Before the advent of the internet and the rise of fibre-optic networks, only very large organisations and firms fully utilised the new technologies. Transnational corporations (TNCs) became the main users of international leased telecommunications lines for Electronic Data Interchange (EDI) and Electronic Fund Transfer (EFT) which are regarded as a key factor for speeding innovation, mobility of capital and competitive advantage within organisations. In general, international EFT encompasses a wide range of payment systems and services differing among others by ownership, user access and geographical extent. For interbank transactions the SWIFT (Society for Worldwide Interbank Financial Communication) network became a cheap, reliable and secure alternative to public services (Langdale, 1994, p. 437). SWIFT is a private international telecommunications service for member banks and qualified participants. It provides an international network for a large range of interbank communications including money transfers, letters of credit and many more. SWIFT was founded in 1973 as a cooperative nonprofit organisation with headquarters in Brussels. In the beginning, it had 239 member banks from 15 countries. Operation started in May 1977 with 15 banks in Belgium, France and Britain. Meanwhile, there are over 7000 members from 194 countries.
For intraorganisational communications, large TNCs and transnational banks alternatively use their own networks with lines leased from PTT (Postal, Telephone and Telegraph) authorities. They are motivated by reliability and availability concerns as well as cost and control considerations (Langdale, 1994, pp. 437 f.). The banks' networks long have become an integral part of competitive strategies to attract large customers with sophisticated transmission requirements such as TNCs, and they are traditionally strongly competing with one another and with institutions like SWIFT, PTTs and others offering EFT services. On the other hand, TNC networks themselves have become a source of competition to financial institutions in handling a substantial share of their own financial transactions bypassing the services of banks. At the beginning of the 1990s internal data and voice transmissions with TNCs comprised an estimated 50 per cent of all cross-border communication flows between nations (Graham and Marvin, 1996, p. 138). The spatial form of those networks linking banks and firms across time zones and spaces may be categorised into three broad types reflecting various degrees of internationalisation and globalisation (Figure 1). In traditional centralised systems (Figure 1a) both regional and national nodes are connected directly with the firm's headquarters or centre of control of international activities by low-speed leased circuits. On the other hand, in a truly global network (Figure 1c) there are linkages between national and regional hubs with the latter connected to the global centre and to one another by high-speed circuits. Regional hub-and-spoke networks (Figure 1b) are an in-between form. TNCs and transnational banks may use either one of the three or a mixture of types. The larger and more extensive their operations, the more likely are they to use either regional hub-and-spoke or global networks.
Retail Banking and Exchanges
The new information technologies profoundly affected customer relations in the financial industry. Traditional modes of business were first altered by automatisation and the possibilities of home banking. The first automated teller machines (ATMs) - unmanned terminals used to dispense cash, take instructions on fund transfers and summarise information on the state of the account and other features - emerged during the 1970s and early 1980s. The greatest investments in this field were made in the United States, followed by France where the nationalised banks were encouraged by government to introduce the new technology. With respect to home banking Scandinavia, and Finland in particular, led the development. This was explained by high labour cost and the isolation of the rural population in those countries (Hamilton, 1986).
Another novelty was credit cards which, again, were first introduced in the USA. This time, it took a full decade before the development reached Europe or Asia. One of the reasons was same-day settlement. The US Fedwire system, a private network for transfers between financial institutions with accounts at the Federal Reserve Bank, was far ahead of comparable systems elsewhere. Founded in 1914, Fedwire was computerised in the early 1970s and modified in 1982. The system handled transfers relating to commercial as well as money market, foreign exchange and securities transactions. In comparison, in the 1970s, the nearest equivalent in Britain, the Bankers' Automated Clearing Service (BACS), founded in 1968, was intended to handle only standardised payment instructions. In London, until 1984 "the actual clearing of cheques between banks was achieved by bringing the cheques, orders and instructions to a central clearing hall ... where they were redistributed to the desks of the clearing banks and the Bank of England." (Hamilton, 1986, p. 36)
The emergence of the internet, a network based on packet-switching technology and independent of command and control centres (Castells, 1996), added an aditional dimension to the world of electronic retail finance. New market segments and structures emerged. Cost reductions by electronic commerce were substantial. One credit card company estimated the cost of processing purchase orders to have declined from $125 to $40. In 2000, cost of a financial transaction for a US bank was $1.27 for a teller, $0.27 for an ATM and $0.01 for an online transaction. Costs in back-office operations and brokerage transactions were reduced, too, leading to online brokerage fees of below $5 compared to those of traditional discount brokers exceeding $50 (Lucking-Reiley and Spulber, 2001, p. 57). But, the most dramatic effect of the internet was the virtualisation of individual customer activities. Suddenly, everyone with a PC and telephone had access to financial information worldwide and trading opportunities never known before. This development was not restricted to the US. There were estimates for Europe in January 2000 that an average of 466 new online accounts were being opened in Sweden every day, 685 in Britain and even 1,178 in Germany. At the same time, in Asia, with around 30 per cent of stock market turnover South Korea had the highest proportion of online trading in the world. And in Japan, the market leader Nomura boasted its online accounts to be growing by 1,100 a day (The Economist, 2000). Banks, securities houses and non-traditional players started to set up trading platforms and offered all kinds of financial services online. Meanwhile, most are facing difficult times since online trading by individuals has ground almost to a halt after the near-collapse of the market in new-economy stocks in the year 2000 and the concomitant sharp decline in profit margins. Pressures to consolidate and find partners are rising, and those pursuing a multi-channel banking approach keeping voice-broking elements beside their electronic systems now seem to fare best (Semler, 2001).
In recent years, both in retail and wholesale trading brokers and exchanges faced increasing competition from so-called electronic communications networks (ECNs). Those include professional trading systems such as Instinet, which has become the world's biggest electronic broker. These days, Instinet is trading in about 40 markets with offices in London, Frankfurt, Paris, Zurich, Hong Kong, Tokyo and Toronto. Other ECNs are order-matching services like Posit and E-Crossnet. Most are owned by traditional market participants or brokers and regulated as investment companies rather than exchanges. The spread of ECNs dates back to US regulatory changes in the mid 1990s authorising non-traditional players to set up trading platforms. There are estimates that in the United States ECNs meanwhile account for 30 per cent of trading in securities listed on NASDAQ. In contrast, in Europe, so far, they play a minor role accounting for less than five per cent of equity trading turnover (OECD, 2001, p. 49). However, there is one area where they have started becoming more important, which is foreign exchange trading. Here ECNs are held partly responsible for the recent decline in trading volumes (Bank for International Settlements, 2000, pp. 98 f.).
Another segment of electronic trading that is strongly growing is in fixed-income securities. Only recently have advances in computer software made it easier to categorise and match types of outstanding bonds, and the internet gave small investors access to information that previously had been the province of broker-dealers and large institutional investors. In 1999, there were 68 electronic systems for institutional trading of bonds, compared with 11 three years before, five of which operating only in Europe. In the US, after consolidation in response to overcapacity, contraction in the internet economy and unsuccessful business models, in 2001, the number declined to 49, while in Europe end of 2001 it was grown to 24 systems (Wiggins, 2001). In addition, electronic trading of bonds by retail investors via the internet is growing rapidly. Nevertheless, so far, fixed-income markets are still behind those in equities (OECD, 2001, p. 50).
In bond and equity markets alike competition between traditional and electronic exchanges and ECNs intensified in recent years enforcing a tendency towards alliances, mergers and pan-European and worldwide 24-hour trading. One example is the merger of the bourses of Paris, Amsterdam and Brussels to form Euronext, which then won the battle for LIFFE in October 2001. Others are the (failed) hostile takeover bid for the London Stock Exchange by OM Gruppen of Sweden, and the (equally failed) plan to create iX by merging the London and Frankfurt stock exchanges and possibly Madrid and Milan. Worldwide alliances include the creation of markets in various countries with local partners using a common technology, a strategy applied, for instance, by NASDAQ, which so far established NASDAQ Europe, NASDAQ Japan and NASDAQ Canada. Another example is the Globex Alliance which, beside the CME and SIMEX, meanwhile includes Euronext, Brazil's Bolsa de Mercadorias & Futuros (BM&F), the MEFF Renta Fija Spanish Exchange for Fixed Income Derivatives and MEFF Renta Variable Spanish Exchange for Equity Derivatives and the Montreal Exchange. In addition, there is a CME-LIFFE Partnership (Figure 2).
In all regions of the world, in electronic stock trading traditional leaders are facing competition from smaller exchanges. In the United States, for example, NASDAQ is challenged by exchanges such as the American Stock Exchange, the Cincinnati Stock Exchange and the Boston Stock Exchange which are all seeking to expand into NASDAQ trading aiming to offer services to securities firms and others that trade NASDAQ stocks (Labate, 2001). In Asia, traditional trading centres such as Tokyo, Singapore and Hong Kong are experiencing competition from electronic trading in smaller places like Korea where, for example, trading volumes on Kofex surged since its start in April 1999 reaching an accumulative four million contracts in January 2001. In Europe, smaller electronic exchanges are struggling for survival in the new financial landscape. Prominent examples are VirtX and Jiway. VirtX is a pan-European stock exchange that was formed in 2001 by a merger of Tradepoint, the UK electronic platform, and the Swiss Stock Exchange blue chips. It specialised on cross-border trade undercutting other exchanges with cheaper fees for trading and settlement. Jiway was launched by OM Gruppen and Morgan Stanley Dean Witter in 2000 as an online cross-border exchange for retail investors. In September 2001, it was bought out by OM in an attempt to cut costs by integrating its exchange operations with those of the OM London Exchange.
Trading, Risk Management and Settlement
IT convergence did not only affect retail banking and exchanges. When informations became available instantaneously, or in "real time", and new computer technologies and software programs allowed to monitor and analyse huge data sets and discover so far unseen relations between them, new opportunities arose in other areas as well. This holds in particular for large banks and institutional investors. Examples are program trading and portfolio insurance: The former is a kind of arbitrage between cash and derivatives markets to exploit minor discrepancies in pricing signalled by computers, the latter is a way in which major institutional investors seek to lock in profits on their portfolios, or hedge against losses, by operations in various markets with trades initiated automatically by computer programs. Another example is options trading. Option prices are calculated according to complex formulas which explains why, before the arrival of respective computer facilities, compared to current volumes trading in those markets was rather modest. The availability of those technologies also paved the way for the rise of so-called hedge funds. Those are privately subscribed funds that take highly leveraged positions. Although having their roots in the United States, where they look back to a long history (Bennett and Shirreff, 1994, p. 29), nowadays, most of them are operating largely unregulated from offshore centres.
But, there were also trades and activities that ceased to exist with the emergence of the new technologies. With real time information of remote markets and prices, and the opportunity to react instantaneously to emerging discrepancies, dealers started adjusting quotations quasi automatically, with the result that, for example, in foreign exchange trading the arbitrageur in the traditional sense of the word has largely disappeared. What is now called arbitrage generally is not aimed at taking advantages of different prices in different markets or regions but at exploiting price differences in time. Dealers buy or sell a financial instrument in the hope to make a profit by reversing the transaction at a time in the not too distant future - which in the interbank market, for instance, usually means later in the day (Reszat, 1997, pp. 60 ff.).
With the new technologies at hand new approaches in financial research emerged, too. Stimulated by latest developments in the natural sciences and encouraged by the advances in computer and information technologies, in the 1990s, scholars of finance began to apply concepts such as chaos theory, fuzzy sets and neural networks, that require the analysis of huge data sets, to their own discipline and developed new techniques for time series analysis of high-frequency data which almost instantly found their way into the research departments of banks and other financial institutions. At the same time, and partly as a result of these efforts, overall portfolio and risk management of those institutions became far more sophisticated. Without the new technologies, the shift in emphasis from the classical asset-and-liabilities management (ALM) in banking business to the value-at-risk (VAR) concept as an international standard for risk management would have been unthinkable.
With the possibility to handle ever growing flows of data clearing and settlement of monetary and financial transactions became easier and more transparent, too. As a result, payment system risk in international banking was reduced substantially. Until recently, most large-value interbank payments and settlements systems worldwide operated on the basis of netting. In the course of the day, the systems were keeping track of the net position of banks which were sending thousands of payment instructions to each other, and at the end of the day the net amounts owed were settled by means of transfers between the participants' accounts at the central bank (Borio and Van den Bergh, 1993). Meanwhile, in most G10 countries real-time gross settlement (RTGS) has been introduced, at least for some financial transactions. There are a number of variants of RTGS systems but common to all of them is that funds transfers are settled individually as soon as the corresponding orders are sent provided that the sending bank has sufficient cover in its account with the central bank.
Beside, there are private sector solutions to reduce payment system risk. Not least with the availability of new technologies banks have begun to improve their back office payments processing, correspondent banking arrangements and risk management procedures. Beyond those efforts, institutes worldwide have started thinking about strategies to jointly limit payment system risks. Some are 'netting' trades bilaterally or on a multilateral basis by pooling their trades in a particular currency and cancelling out offsetting ones with settlement at the end of the day. In August 1995, a group of European banks consisting of large British banks as well as several French, Dutch and Scandinavian institutes launched the Exchange Clearing House (Echo) in London thereby extending the concept of netting to a multilateral basis. Participants of Echo no longer make payments to each other but to the Echo clearing house. Compared to bilateral procedures multilateral netting reduces payment flows still further since transfers are made by all members to a single counterparty. A competing system which received regulatory approval in December 1996 is Multinet established by six Canadian and two US banks.
While settlement of interbank payments became safer, in other markets progress was much slower. This holds in particular for cross-border securities settlement, although with considerable differences between regions and markets. For instance, in Europe, the annual costs of maintaining a share clearing and settlement infrastructure amounts to between $1 billion and $1.2 billion, compared to an estimated $600 million in the United States. While for fixed-income securities the emergence of an international bond market in the 1970s led to the creation of two international clearers - Euroclear and Cedel - clearing and settling shares traded between European countries remained strongly fragmented. There are between 20 to 30 institutions in charge of it, some run as for-profit organisations, others not. But, in recent years, with mounting competitive pressures the industry has seen some consolidation in these markets and between bonds and equities markets: Cedel and Deutsche Börse Clearing merged to form Clearstream. Sicovam, the Paris settlement system merged with Euroclear, which was then joined by CIK and Necigef, the central securities depositaries of Belgium and the Netherlands, and in London the CCP, a central counterparty for stocks, has formed as a joint initiative by the London Stock Exchange, the London Clearing House and Crest. End of 2001, the LSE reached an agreement with the London Clearing House and Euroclear allowing its international customers to use Euroclear to settle trades. There is an initiative to create a single central clearing counterparty for Europe's equities markets by the world's biggest banks, but obstacles such as legal and tax differences, cultural and linguistic problems and differences in national supervisory systems are high.
The consequences the new technologies had on the financial industry as a whole and on location choices were far-reaching and not clear cut: On the one hand, they facilitated activities such as data processing, trading, risk monitoring and settlement and, in allowing to steer operations from afar, loosened the ties to particular places. On the other, they made management operations and control much more difficult calling for a growing centralisation of headquarter functions which, in turn, strengthened the importance of location. Beside, they affected firms' spatial behaviour in enabling a rising number of financial institutions to strive for market shares beyond regional borders or on a global level, or to become organised in worldwide networks in order to achieve scale economies through cooperation with others. The result was an increasing overall level of international financial activities, and a rising importance of financial services for the growth and development of cities. Competition between places rose and, at the same time, complementary linkages between them became more pronounced as well. Exchanges merged, but they did not centralise functions to an extent that made them settle down in one single location. Banks started exploiting the comparative advantages of places by buying local firms and concentrating activities like investment banking and foreign exchange trading in one place, but, at the same time, they maintained other functions and regional headquarters and control centres elsewhere. Seen as a whole, spatial relations and structures in international finance became more flexible and complex. This explains why the effects of the new technologies on world financial landscapes attract a growing attention in the literature.
WORLD CITIES IN THE AGE OF ELECTRONIC FINANCE
The advent of the internet, e-mail and electronic trading platforms is often said to have opened a new dimension to the idea of location in the financial industry. Even within firms physical closeness seems no longer necessary and "virtual" market places are increasingly competing with real ones. The first IT revolution with the emergence of the telegraph and telephone largely contributed to the rise and economic prosperity of world cities through its impact on international finance. Now there are growing fears that the second one, characterised by the convergence of communications and computer technologies, may equally contribute to diminishing the cities' future role and reducing their growth prospects. In order to assess whether these fears are justified, in principle, answers have to be found to two fundamental questions, namely, what are the determinants of the world's financial landscapes in general and in which way do the new technologies alter the rationale for financial institutions to choose a particular location?
Centres and Peripheries
Historically, the world's financial landscapes are characterised by patterns of centres and peripheries. In the Middle Ages, Venice and Genoa in the south, and the Champagne fairs, and later Bruges and Amsterdam, in the north were the first nodes of financial activity in Europe. Later, London became the financial centre, first of Europe and then of the world, which kept this role until the first World War, then temporarily lost it to New York, only to regain it in the 1960s with the advent of the euromarkets (Reszat, 2000). But, there are regional differences. For example, in the New World the first financial centres emerged only in the 19th century. In Asia, the phenomenon is still much younger although the region is the origin of one of the oldest informal remittance and money exchange systems (Reszat, 2001). In contrast to western banks, money changers and bankers in Asia traditionally relied on a private web of finance long having no need for public institutions such as fairs and exchanges functioning as a "centre". Beside, it is not always international finance that determined a place's rise as a financial centre in the first place. Both, Tokyo and New York, for example, owe their importance for the financial industry more to their countries' economic strength, and their regional and worldwide economic dominance, than to the international activities concentrated there. Regional differences also exist with respect to the dynamics of competition between centres. While in Europe, in the process of monetary union, London's predominance is challenged by Frankfurt and Paris, and in Asia places like Singapore and Hong Kong are threatening Tokyo's leading role - with new competitors on the horizon in both regions -, in the United States Chicago is still the only serious rival to New York.
Traditionally, regional and international financial centres are located in world cities. Those are characterised by high densities of economic factors such as labour, money, commodities, services and information. Their influence is stretching far beyond their boundaries to places and regions worldwide. Together they form an international network that, although encompassing only a small fraction of the earth's surface and population, represents a large share of the world's total production and consumption. World cities are large urbanised regions that are characterised by dense patterns of interaction rather than by political-administrative boundaries. In recent years, the growing connections between them have influenced a new thinking about global economic activities focusing on nodes organising flows and inter-city linkages rather than nation states (Beaverstock et al., 2000; Taylor, 2001). Serving as control centres of the global system they can be arranged into a hierarchy according to the economic power they represent - at least, with respect to their global financial articulation and as far as the top is concerned where New York, London and Tokyo are the undisputed leaders (Friedman, 1995; Taylor et al., 2002).
Traditional Determinants of Location
What makes financial institutions locate in centres? In economics, first approaches to answer this question can be found in the concept of the new economic geography with its emphasis on the interplay of centripetal and centrifugal forces as determinants of agglomeration (Reszat, 1999). In principle, the concept dates back to the theory of central places developed by Walter Christaller and August Lösch in the 1930s and 1940s which was explaining firms' location decisions as a tradeoff between scale economies and transportation costs. The new economic geography is further developing this idea in stressing the role of interdependence and interaction. Here, a firm's decision on where to locate depends on all other firms' choices. On the one hand, firms like having other businesses nearby because those help attracting customers to an area and add to the variety of local services offered. These advantages are called centripetal forces. On the other hand, they dislike the proximity of others because they compete for customers, land and workers. These motives are the centrifugal forces driving them away from one another. Then, the interplay of both centripetal and centrifugal forces determines the process of business migration and the forming pattern of agglomeration (Krugman et al., 1999).
Financial institutions partly differ from other firms in their motives for choosing a particular location. In former times, the need for physical money transfers between banks and the proximity of the stock exchange played a dominant role. With developing information and communication technologies other arguments such as trading frictions and transactions costs became more important. In the literature, among the centripetal forces named as relevant in this context are economies of scale in the payment mechanism, informational spillovers and market liquidity, among the centrifugal ones market access costs, policy intervention and localised information that cannot be traded globally (Gehrig, 1998). Other arguments relate to the effects of congestion such as high rents and to socio-institutional and cultural factors (Thrift, 1994, Grote et al., 1999, Reszat, 2000).
The study of centripetal and centrifugal forces helps understand why banks tend to form "clusters". It does not necessarily explain why they locate in world cities. Considering the environment in which large international financial firms are operating arguments why they would wish to be in the world's big centres can be broadly summarised under three headings. Those are risk management, dependence on producer services and innovation. Among those, reducing risk and uncertainty perhaps has the highest priority. Uncertainty is particularly high when internal and external relations, operations and management tasks are complex, the numbers of products sold, and of countries in which they are sold, are high and personnel and financial resources are committed to many markets with different cultural, political, social or regulatory environments. In these cases, location decisions are strongly influenced by the need to closely monitor activities. Experience has shown that human management of the risks financial institutions run cannot be entirely replaced by electronic surveillance. This does not only explain why firms search to have the riskiest parts of their operations under one roof or, at least, nearby (Edwards, 1998). It is also the reason why they tend to locate in those places that will give them access to as much information as possible, be it through contact with local businesses or by allowing them to monitor the behaviour of competitors. World cities then are an obvious choice. As Daniels (1993, p. 114) puts it: "Shadowing competitors may seem an unlikely way to capture a share of new markets but, in some services such as banking or finance, pioneering location choices are not made lightly because of the adverse signals that they may be perceived to convey to investors or businesses."
Following Daniels, there are influences of location decisions that can be categorised into urbanisation economies and localisation economies. The former include benefits from access to infrastructure of transport, telecommunications, housing and office buildings and the availability of large and diverse labour markets. The latter refer to the benefits of proximity to similar services or other economic activities either as input providers or clients. World cities offer a wide range of so-called "producer services" provided by specialised firms of which international financial institutions with large and growing geographic and functional dispersion of activities have become more and more dependent (Sassen, 1991; 1995). There are information processing services like marketing, insurance, accountancy, property management, advertising and legal services. Others are goods-related services such as distribution, transport management, infrastructure maintenance and installation as well as repair and maintenance of communications equipment. A third category is personnel support services including welfare, catering or personal travel and accomodation (Daniels, 1993, p. 26). The firms offering these services often have rather limited location choices because, operating under high competitive pressures, they depend on having access to the best possible information, labour and specialist advice which are most available in the big cities and this, in turn, is attracting their clients to these places, too.
But, risk reduction and dependence on producer services are not the only explanations for the attractiveness of world cities to financial institutions. Another key factor is innovation and networking. Traditionally, cities are seen as seedbed for innovations. The city as "creative milieu", as "networked society" where the proximity of advanced technology, high-quality education and sophisticated finance is generating new ways of economic organisation, new forms of production and new industries and wealth providing a basis for patronage of arts and culture that, in turn, are again stimulating knowledge and technology is a recurring theme in history (Hall, 1998). Another recurring theme is the role of outsiders coming into the city and creating something new which then becomes a source of economic progress and growth. Traditionally, world cities are places where many cultures are coming together, where the local and the global are mixing and often mutating to something wholly different. "It is in the juxtaposition, mutations and connections of different cultural spaces, in the overlaying of contradictory cultural landscapes over each other that creativity and vitality may emerge." (Homi Bhabha cited in Crang, 1998, p. 175).
In the history of western finance foreign bankers always played a prominent role in contributing to the big centres' creative milieux. Technical progress, economic growth, city development and financial innovations often went together stimulating one another. Examples are found in the environment in which finance and insurance of long-distance trade blossomed in Europe since the merchant empires of the Middle Ages, in Amsterdam which was a magnet for both science and the arts and international capital in the 16th and 17th centuries, in the creation of debt instruments in 19th century Paris to finance Haussmann's city building projects, or the invention of bond houses during the New York boom in skyscrapers and office development in the mid-1920s - to name only a few. For a long time, it was taken for granted that among all human activities this interplay of innovations and the creative milieu in which they flower were particularly depending on face-to-face contact. But, with the latest developments in IT technology this dependence has become at least arguable and for a while it seemed as if the emergence of technopoles and large industrial complexes ranging from Silicon Valley over Akademgorodok in Siberia to Tsukuba, Japan, would indicate that the role of cities as seedbed for innovations had undergone substantial changes reducing their importance in the world economy.
Cities in the Geography of Information Technology
A closer look at the overall position of cities in the geography of information technology is putting this impression into perspective. These days, it is widely accepted to talk about the "dissolution of cities". This has become one of the grand metaphors of the influence of the IT revolution (Graham and Marvin, 1996, p. 8), although the argument is not new: The economic decline of locations due to electronic communications, the advance of personal computing and related developments eliminating the need for cities as centres of interaction has been conjured up since the early 1970s (Moss and Townsend, 1998). But, studying, for example, the visualisation of internet traffic flows in the 1990s (Mapa Mundi, 2002), transatlantic submarine cable linkages (Telegeography, 2002) or other indicators of worldwide information activities gives the overall impression that global cities somehow managed to redefine their role in the networks and geography of the new information technologies.
Recent research confirms that the spatial distribution of those technologies shows familiar characteristics (Gorman, 2001). Domains and connectivity cluster predominantly in big cities. The large metropolitan areas of the industrialised world have the highest concentration of internet domains and, at the same time, are the gateways to internet connections in other world regions. For example, in the United States, the 20 most networked cities account for 4.3 per cent of the nation's population, but 17.1 per cent of its domains (Moss et al., 1997). Apparently, those new electronic grids spanning the globe and being capable of going anywhere remain inherently nodal. As Graham and Marvin (1996, p. 3) wrote: "Urban areas are the dominant centres of demand for telecommunications and the nerve centres of the electronic grids that radiate from them", and financial services make for a decisive part of the traffic flooding across those networks and play a dominant role in linking the nodes. A look at the top five most networked cities in the United States shows that Manhattan has by far the highest number of domains (Table 1), Singapore and Hong Kong have become critical nodes in an Asian-Pacific network geography, and in Europe, after far-reaching deregulation in the 1980s, London developed into a centre for all kinds of information-rich service industries ranging from finance over broadcasting and publishing to advertising and many more (Dodge and Kitchin, 2001, p. 48).
Much of the discussion about the implications of IT technology for the financial industry, for the future role of cities and the competition of places focuses on the growing complexity of management operations and control that is reinforcing the need for centrality. But, for the cities it is not necessarily headquarters and control functions that matter - for example, many investment banks in London have their headquarters located elsewhere - but a vast spectrum of financial services and activities. Those activities account for large numbers of people living in a place, paying taxes, consuming, using its amenities and in contributing to the place's "crowdedness" enhance its attractiveness to others searching the company of like-minded. With the rise of IT technologies, their numbers are in decline. Even with headquarters staying in the central business districts of the world's metropoles, in other areas of finance, there is an observed tendency towards dispersion. New technologies have reduced the need for proximity in sectors not affected before, such as customer relations and retail business. Call centres, "software complexes" and internet suppliers with locations far away from the world's financial metropoles blossom, and, while leaving high-order activities in urban agglomerations, financial institutions are increasingly shifting part of their low-order activities such as back-office operations to other places or outer parts with cheaper labour and lower rents thereby diminishing the role of centres and central business districts.
Does this mean "the end of cities"? A closer look at the underlying nature of the phenomenon of virtualisation and its implications for the world's financial landscapes may help exploring the possible extent of these tendencies providing some insights into the general mechanisms at work. What characterises virtual activities and in which way do they differ from traditional ones? What is lost and won in the process of virtualisation? To what extent can electronic flows replace physical activities, and to what extent and in which ways do they complement them in the world of finance?
Spaces and Places
In general, the study of real and virtual market places immedeately raises the question of how to define "place", of what kind of experiences places provide and in which way those differ from their virtual counterparts. In disciplines such as sociology, cultural geography and architectural studies, the basic idea of "place" is the sense of belonging it conveys to people beyond any simple concept of location. Places differ from one another in the set of cultural characteristics they show, the patterns of behaviour and interaction developing there in course of time, and the shared experiences they mean for people living there. People tend to define themselves through a sense of place. When asked, they emphasise being "a Londoner" or "a New Yorker". Financial communities do not differ in this respect from other groups of society. They, too, identify themselves with the places they live in. Those working in metropoles like London and New York are well aware of their special status experiencing a certain way of life and the feeling of being part of a special culture (Reszat, 2000). Trust and familiarity are built by common ideas, behaviour patterns, norms and "rituals", and, through close contacts from constantly trading with one another and other concomittants of the globalisation process this feeling is in parts transmitted to financial communities in other places engaged in the same kinds of activities. Influences like these establish non-measurable location advantages that nonetheless do exist. Whether similar advantages are enjoyed by participants of purely "virtual" networks and communities, lacking the experience of a physical environment is not easy to see.
Place in this sense differs from "space" by the evolution of shared experience, by having a past and future "that binds people together round them." (Crang, 1998, p. 103). The term "cyberspace" indicates that this characteristic is widely regarded as lacking in virtual relations. Virtualisation is often equalised with the erosion of place, extinguishing all specifities that matter to people. Lives are considered to become somehow diminished and reduced in their traditional dimensions with the computer screen as one of Marc Augé's "non-places" like airports, motorways and hotels where neither identity, nor relations, nor history exist (Augé, 1995). Individuals in those spaces are facing a kind of "contractual solitariness" (Crang, 1998). Alone, or in small groups, they are related to wider society through very limited and specific interactions. They are linked to their surroundings mainly through words and texts, through instructions - be they prescriptive, prohibitive or merely informative - supported by signboards, screens and posters (Augé, 1995, p. 96).
On the other hand, virtual reality itself opens new dimensions. For example, Crang et al. (1999) distinguish between four kinds of experiences. They study virtuality as simulation, as the "other" in relation, and in opposition, to the real with both depending on one another; virtuality as complexity, which refers to the multiple configurations of self and world constantly constructed and reconstructed developing characteristics of self-organisation and emergent order; virtuality as mediasation, as one stage in the history of "mediated and distanciated communication" (Crang et al., 1999, p. 10); and virtuality as spatial emphasising the "other" geography of cyberspace with its relation to existing social, political and economic geographies. In this context, the realm of international finance is, above all, concerned with two effects. First, the extent to which virtual reality replaces, or complements, face-to-face interaction. Second, the way in which virtual reality leads to an accelerated transformation, a kind of "tunnel effect" or "warping" of time and space barriers (Graham and Marvin, 1996, p. 60). Both affect the prospects of world financial landscapes.
One crucial characteristic of face-to-face interaction is that it allows tacit bargaining and communication where each actor watches and interprets others' behaviour, fully aware that his own decisions are being carefully watched, interpreted and anticipated, too (Schelling, 1980). Traditionally, tacit communication is an essential part of the financial business which matters for some activities more than for others. As a consequence, in some cases it is also easier to be replaced by other forms of communication. For example, for trading securities in bulk, making interbank payments or in standardised foreign exchange dealing the advantages may be minuscule or even non-existent. For mergers and acquisitions, the management of investors' portfolios or the lead management of syndicates they are high. Tacit communication is an essential element in building trust which, in turn, is a major prerequisite for many financial transactions. In these cases, the impossibility of face-to-face contact is regarded as connected with a loss of "authenticity". A virtual market place is said to be no longer an "authentic" place and therefore no substitute for a real city centre.
What is won in exchange for authenticity is an acceleration of transformations or "tunnel effect" resulting in a wider reach and greater speed and efficiency. In traditional geographies people would experience spaces and places change slowly in the course of history with the hierarchy of cities, their rise and decline, undergoing slow changes as well (Braudel, 1986; Hohenberg and Lees, 1996). Under virtual reality people experience that barriers of time and space are no longer of much importance. Cities become linked together into networks, with the spaces in between largely excluded and denied access. Market places are no longer perceived as local phenomena, local entities as parts of cities. For a wide range of financial activities and instruments London, New York and Tokyo represent one single global market. One consequence is that people are able to be present in different markets and localities at the same time. Another is that cities in these worldwide networks may be dominating in one area such as finance while at the same time being inferior in other realms so that their relations are no longer determined by a strict hierarchy but by the evolving interdependencies (Graham and Martin, 1996, p. 61). And, to the extent that electronic networks and linkages can be compared to, and are able to replace, traditional "networked societies" within cities they may well contribute to the decline of central places.
An Experiential Continuum
Talks of the "end of cities" as hubs of financial activities in the electronic age usually are based on the view of a complete separation of virtual and real activities. But, as a closer look reveals, this assumption does not hold. On the contrary, cyberspace and geographic space are closely intervowen. These days, there is no part of the financial industry that is doing without the new technologies. The bank manager preparing for a talk with a client relies on virtual networks and services for data and information from all parts of the world before entering the meeting presenting the arguments as convincingly as possible. The brokers that fared best in the recent consolidation in e-business are the ones that kept voice-broking elements beside electronic systems. The investor in search of hints to future profit opportunities from tacit communication will turn to the computer and, afterwards, confront his counterpart with the latest news or results of research observing the reactions. And when both have finished their business they will go back to the screen feeding in their latest findings in order to analyse and combine them with other information laying the foundations for future decisions. On the other hand, even the age of virtual business cannot do without personal communication over large distances, and in the world of finance the jet aircraft still plays an indispensable role. The financial community worldwide is part of an international class of people brought up by globalisation. One characteristic of those people is their constantly being in reach of, and contact with, firms, clients and one another by computer, mobile telephone and beeper (Micklethwait and Wooldridge, 2000). The bank manager in the airport lounge with a laptop on the knees is a familiar sight.
In practice, there is no virtuality without corresponding reality, and depending on context the virtual market place is rather complementing than replacing real financial activity. Identities "explored and acted out online are always contextualised within experiences offline. ... Conversely, our lives offline become embodied through our memories and experiences online, so that a recursive process exists as the virtual is realised and the real virtualised." (Dodge and Kitchin, 2001, p. 24) The financial world seen in this way is an "experiential continuum", as Dodge and Kitchin name it, between the materiality of geographic space and the virtuality of cyberspace. This calls for a revision of the virtualisation effects mentioned earlier:
On the one hand, the loss of authenticity in virtual market places is not as complete as at first view, and even "authentic" trust building and trust requiring activities demanding personal contacts are surrounded and supported by virtual realities. On the other hand, the acceleration of transformations in virtual reality is slowed down by interaction with the real world. Financial deals may be done surpassing time and space, but in the place they are invented and decided, and in the "switches" from the virtual to the real, and back again, the traditional laws of physics - and psychology - still hold. Financial risks may be calculated by sophisticated computer programs. But, the time it takes to digest the results and develop a strategy in response is limited by human abilities and deficiencies. As a rule, firm managers are well aware that each trial to shorten the process of personal communications and investigations needed to find out what really were behind the numbers, or to cope with the remaining uncertainties from far away, may prove a fatal decision.
What are the consequences of this intervowenness of real and virtual for the role of cities? To the extent that virtual activities cannot be separated from their real surroundings, the traditional arguments for and against centralisation still hold setting clear limits to the diffusion process. As a result, even in the age of virtuality cities continue to compete with - and complement - one another as financial centres. Within cities, to the extent that the new technologies allow to shift low-order activities to outer places, the process is determined by traditional arguments such as rents and labour costs and the scarcity of space in cities' central business districts. But, experience shows that, typically, even in these cases, there is a tendency to locate in suburbia rather than in distant regions which has both internal and external reasons. In firms' internal relations, one explanation is risk considerations and the experience that electronic surveillance is no substitute for human management. In external relations, it is above all trust building activities, the need to become familiar with a special context, innovation and networking. In general, the less financial activities require spatial proximity in order to manage risks, gain access to information, establish social contacts or get impulses from the creative milieux of local communities, the more cost arguments matter providing a reason for dispersion, at least on a limited scale.
Thus, the strongest foreseeable impact of information technology is on the spatial organisation within cities. This is reinforced by the expansion of workplaces and spatial requirements of "intelligent buildings" for which traditional financial districts often are lacking space and which, therefore, in parts have to locate outside the old centres. La Defense in Paris and London's Docklands are two examples. But, here again are regional differences. In the United States, the centres in major cities such as New York and Chicago have been rebuilt many times without caring much of urban infrastructure and design, and there are vast empty spaces left for rebuilding according to the requirements of technology. In contrast, in Europe, urban centres are much more protected and rarely contain significant stretches of abandoned space. Here, the financial district often has become less compact extending more into a metropolitan area in the form of a grid of nodes of business activities (Sassen, 1995).
From the role of IT in the history of finance, its very nature and its relation to world city growth, several conclusions can be drawn. First, unlike in many other sectors, in international financial relations electronic information transmission, data processing and trading is not a new phenomenon. The "internet revolution" here brought rather a gradual change. Financial services as the forerunners of globalisation have a long tradition of using advanced information and communication technologies for overcoming the limits of time and space. Their formation of "clusters" and location in centres largely contributed to rise and economic prosperity of cities. Second, even in the age of electronic connectedness and virtual finance location still matters and there are no signs that the bulk of financial services will shift away from the world's metropoles. The electronic grids of financial institutions spanning the globe are inherently nodal, and the cities so far managed to redefine their role as nodes in the networks and geography of the new technologies. Third, the myth of the "dissolution" of cities is based on the assumption of a perfect separation of virtual and real activities that, at a closer look, does not hold. Financial decisions are made in an "experiential continuum" between the materiality of geographic space and the virtuality of cyberspace. Neither the loss of authenticity nor the acceleration of transformations as a result of virtual reality are complete. Virtual markets and processes complement rather than replace existing real ones. Fourth, the biggest impact of the new technologies so far is on the shape and spatial organisation within cities. Technological progress allowed financial institutions to shift parts of activities to suburbia in face of rising costs, and the lack of space meeting the requirements of an extended workforce and "intelligent buildings" have led to a spread beyond old city centres. As a consequence, cities' financial districts appear less compact than in former times, but there are limits to the diffusion process since many activities continue to require proximity.
New technologies are characterised by the fact that nowbody ever knows exactly where they will lead. In the middle of the 19th century, one could perhaps foresee that the railways would change the geography of countries, "but no one anticipated the simple device of the commuter ticket which would allow suburbs to spread and finally turn cities inside out" (Hall, 1998, p. 943). However, the world of finance appears different in that, as the preceding chapters have shown, the history of information technology in the financial services industry is a far more evolutionary and less transforming process. In the literature, there is a tendency to overstate the importance of the phenomenon. But, as one author puts it: " ... associations of technologies with modernity are contingent not only historically but also geographically; for much, indeed most of the world, the telephone is still thoroughly new and modern ..." (Crang et al., 1999, p. 3). In this sense, cities remain exceptional places, and the financial industry requires the infrastructure and environment provided by these places to prosper.
Augé, M. (1995) Non-places - Introduction to an Anthropology of Supermodernity. London: Verso.
Bank for International Settlements (2000) 70th Annual Report. Basle, 5 June.
Beaverstock, J.V., Lorimer, H., Smith, R.G., Taylor, P.J. and Walker, D.R.F. (2000) Globalization and World Cities: Measurement Methodologies, Applied Geography, 20 (1), pp. 43-63.
Bennett, R. and Shirreff, D. (1994) Let's Bash the Hegde Funds, Euromoney, April, pp. 26-33.
Borio, C.E.V. and Van den Bergh, P. (1993) The Nature and Management of Payment System Risks: An International Perspective. BIS Economic Paper, No. 36, Basle: Bank for International Settlements.
Braudel, F. (1986) Sozialgeschichte des 15.-18. Jahrhunderts, Vol. 3: Aufbruch zur Weltwirtschaft. München: Kindler.
Castells, M. (1996) The Rise of the Network Society. Oxford: Blackwell.
Crang, M. (1998) Cultural Geography. London: Routledge.
Crang, M., Crang, P. and May, J. (1999) Introduction, in M. Crang, P. Crang and J. May (eds) Virtual Geographies - Bodies, Space and Relations, London: Routledge, pp. 1-20.
Daniels, P.W. (1993) Service Industries in the World Economy. Oxford: Blackwell.
Davies, S. (1998) Success Masks Market Turmoil, The Financial Times, 24 March.
Deutsche Telekom AG (2001) Neue transatlantische Rennstrecke Europa-USA, press release, 21 March.
Dicken, P. (1998) Global Shift. London: Paul Chapman Publishing Ltd.
Dodge, M. and Kitchin, R. (2001) Mapping Cyberspace. London: Routledge.
Dubofsky, D.A. (1992) Options and Financial Futures. New York: McGraw-Hill.
Dufey, G. and Giddy, I. (1994) The International Money Market. Englewood Cliffs, NJ: Prentice Hall.
The Economist (2000) Going for Brokers, 20 May.
The Economist (2001) Enter the little Guy, 14 April.
Edwards, B. (1998) Capitals of Capital, A Survey of Financial Centres, The Economist, 9 May.
Favier, J. (1992) Gold und Gewürze - Der Aufstieg des Kaufmanns im Mittelalter. Hamburg: Junius.
Ferguson, N. (1999) The House of Rothschild - The World's Banker 1849-1999. New York: Viking.
Friedman, 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: Cambride University Press, pp. 21-47.
Garbade, K.D. and Silber, W.L. (1994) Technology, Communication and the Performance of Financial Markets: 1840-1975, The Journal of Finance, 33 (3), June 1978, reprinted in: R. Roberts (ed.): International Financial Centres - Concepts, Developments and Dynamics, Vol.1, Aldershot: Edward Elgar, pp. 446-459.
Gehrig, T. (1998) Cities and the Geography of Financial Centers, mimeo, Albert-Ludwigs-Universität Freiburg, March.
Gorman, S. (2001) Where are the Web Factories? The Urban Bias of e-Business Location, internet: http://www.informationcity.org/research/web-factories/Gorman-webfactoriespaper.pdf, February.
Graham, S. and Marvin, S. (1996) Telecommunications and the City. London: Routledge.
Grote, M. H., Harrschar-Ehrnborg, S. and Lo, V. (1999) Technologies and Proximities: Frankfurt's New Role in the European Financial System, Johann Wolfgang Goethe-Universität Working Paper Series on Finance and Accounting, No. 46, Frankfurt, December.
Hall, P. and Preston, P. (1988) The Carrier Wave. London: Unwin Hyman.
Hall, P. (1998) Cities in Civilization. London: Weidenfeld & Nicolson.
Hamilton, A. (1986) The Financial Revolution - The Big Bang and the Explosion of the World's Money Markets. Harmondsworth: Penguin.
Headrick, D. R. (1988) The Tentacles of Progress, Oxford: Oxford University Press.
Hohenberg, P.M. and Lees, L.H. (1996) The Making of Urban Europe 1000-1994. Cambridge, MA: Harvard University Press.
Holzmann, G. J. and Pehrson, B. (1994) The Early History of Data Networks, internet: http://www.it.kth.se/docs/early_net/.
Inwood, S. (1998) A History of London. London: Macmillan.
Keenan, W. M. (1987) The Securities Industry: Securities Trading and Investment Banking, in: E.I. Altman (ed.) Handbook of Financial Markets and Institutions, New York: John Whiley & Sons.
Kindleberger, C. P. (1993) A Financial History of Western Europe. Oxford: Oxford University Press.
Krugman, P., Fujita, M. and Venables, A. J. (1999) The Spatial Economy. Cambridge: The MIT Press.
Labate, J. (2001) Nasdaq Faces Challenges from Smaller Exchanges, Financial Times, 7 November.
Langdale, J. (1994) Electronic Funds Transfers and the Internationalisation of the Banking and Finance Industry, Geoforum, 16 (1), 1985, reprinted in: R. Roberts (ed.) International Financial Centres - Concepts, Developments and Dynamics, Vol.1, Aldershot: Edward Elgar, pp. 433-45.
Lucking-Reiley, D. and Spulber, D.F. (2001) Business-to-Business Electronic Commerce, Journal of Economic Perspectives, 15 (1), pp. 55-68.
Mappa Mundi (2002) internet: http://mappa.mundi.net/maps/maps_008/arc_jan.html.
McGahey et al. (1994) Financial Services, Financial Centers: Public Policy and the Competition for Markets, Firms, and Jobs, excerpt, reprinted in: R. Roberts (ed.) International Financial Centres - Concepts, Developments and Dynamics, Vol.1, Aldershot: Edward Elgar, pp. 95-156.
Micklethwait, J. and Wooldridge, A. (2000) Piloted by a New Ruling Class?, Financial Times, 27/28 May.
Mokyr, J. (1990) The Lever of Riches. Oxford: Oxford University Press.
Moss, M.L., Townsend, A. and Wagner, R. F. (1997) Manhattan Leads the 'Net Nation, internet: http://urban.nyu.edu/archives/net-nation/net-nation.pdf, August.
Moss, M.L. and Townsend, A.M. (1998) How Telecommunications is Transforming Urban Spaces, internet: http://www.informationcity.org/.
O'Connor, A. (2001) Reuters Ready to Press Ahead with IPO of Broker, The Financial Times, 4 April.
OECD (2001) Future Prospects for National Financial Markets and Trading Centres, Financial Market Trends, No. 78, March, pp. 37-72.
Reed, H.C. (1994) Appraising Corporate Investment Policy: A Financial Center Theory of Foreign Direct Investment, in: C.B. Kindleberger, D. Audretsch (eds) The Multinational Corporation in the 1980s, reprinted in: R. Roberts (ed.) International Financial Centres - Concepts, Developments and Dynamics, Vol.1, Aldershot: Edward Elgar, pp. 55-80.
Rennie, J. (2001) The First Optical Internet, Scientific American, January.
Reszat, B. (1997) The Japanese Foreign Exchange Market. London: Routledge.
Reszat, B. (1999) Emerging Financial Centres, Self-Organisation and Evolution, in Homo Oeconomicus, 15 (4), pp. 459-481.
Reszat, B. (2000) Centripetal Forces in European Financial Markets - Cultural Identity and Collective Memory as Determinants of Financial Agglomeration, paper presented at the HWWA-Conference on "Glocalisation in Europe", Hamburg, May 3-5.
Reszat, B. (2001) Hawala, internet: http://www.hwwa.de/Projekte/IuD_Schwerpunkte/IDSPs/Asia_Gateway/Hawala.htm.
Ridding, J. and Montagnon, P. (1998) Asia Downturn Reignites Rivalries between HK and Singapore, The Financial Times, 25 March.
Sassen, S. (1991) The Global City - New York, London, Tokyo. Princeton: Princeton University Press.
Sassen, S. (1995) On Concentration and Centrality in the Global City, in P.L. Knox and P.J. Taylor (eds): World Cities in a World System, Cambridge: Cambride University Press, pp. 63-75.
Sassen, S. (1999) Global Financial Centers, Foreign Affairs, 78 (1), pp. 75-87.
Schelling, T.C. (1980) The Strategy of Conflict. Cambridge, MA: Harvard University Press.
Semler, P.K. (2001) Fineco Overtakes European Online Broking Rivals, Euromoney, August, pp. 86-89.
Standage, T. (2000) The Victorian Internet. London: Phoenix.
Strouse, J. (1999) Morgan - American Financier. New York: Random House.
Taylor, P.J. (2001) Specification of the World City Network, Geographical Analysis, 33 (2), pp. 181-194.
Taylor, P.J., Walker, D.R.F., Catalano, G. and Hoyler, M. (2002) Diversity and Power in the World City Network, GaWC Research Bulletin, 56, internet: http://www.lboro.ac.uk/gawc/rb/rb56.html.
Telegeography (2002), internet: http://www.telegeography.com/products/maps/cable/index.html.
Thrift, N. (1994) On the Social and Cultural Determinants of International Financial Centres: the Case of the City of London, in S. Corbridge, R. Martin and N. Thrift (eds) Money, Power and Space, Oxford: Blackwell, pp. 327-55.
Van Duyn, A. (2001) Only the Best Will Survive: Electronic Bond Trading, The Financial Times, 28 March.
Wiggins, J. (2001) Electronic Trading Systems Diminish, The Financial Times, 13 December.
Williams, K.L. (2000) Price Discovery in Multiple-Dealer Markets: The Case of the Interbank Foreign Exchange Market, Claremont Colleges working paper in economics, August.
Winston, B. (1998) Media Technology and Society, A History: From the Telegraph to the Internet. London: Routledge.
Table 1: The Top Five Most Networked Cities in the United States
Source: Moss et al. (1997), Figure 3.
Figure 1:Types of International Telecommunications Networks for a United States TNC
a) Centralised Network
b) Regional Hub-and-Spoke Network
c) Global Network
Source: Based on Dicken 1998: Figure 5.7.
Figure 2: Exchanges and Alliances
Edited and posted on the web on 26th February 2002