1 Introduction: The Telegraph as Financial Infrastructure

Few technologies changed financial markets more profoundly than the telegraph. Before the telegraph, information and capital moved between financial markets only at the speed of sailing ships or the overland mail. Pricing disequilibria between markets, and price volatility within them, persisted for significant periods of time (Koudijs, Reference Koudijs2016). The introduction of the telegraph to financial markets in the mid-nineteenth century changed this. Financiers rapidly operationalized the telegraph and used it to pursue much faster arbitrage operations between markets (Garbade and Silber, Reference Garbade and Silber1978; Michie, Reference Michie1997; Hoag, Reference Hoag2006; Müller and Tworek, Reference Müller and Tworek2015). Thanks in large part to the global spread of the telegraph, financial markets at the end of the nineteenth century were more integrated than ever before (Neal, Reference Neal1992; Bordo, Eichengreen, and Kim, Reference Bordo, Eichengreen and Kim1998; Campbell and Rodgers, Reference Campbell-Verduyn2017). Beyond enabling securities arbitrage, the telegraph also enabled money orders and remittances to be transmitted between both individuals and financial institutions (Magee and Thompson, Reference Magee and Thompson2006). Telegraphic infrastructure cast a long shadow in the financial system. When the telegraph was replaced by the telex in the mid-twentieth century, the system continued to operate using telegraphic cables laid in the 19th and early 20th centuries (Scott and Zachariadis, Reference Scott and Zachariadis2014). In the late twentieth century, when fiber optic cables finally replaced telegraph cables, the new cables followed the undersea routes first established by telegraph companies a century before (Eichengreen, Lafarguette, and Mehl, Reference Eichengreen, Lafarguette and Mehl2016). By enabling financial actors to move capital and information at far greater speeds across far greater distances, the telegraph helped usher in the first age of financial globalization and has no small claim to being one of the most impactful technologies in the history of finance (O’Rourke and Williamson, Reference O’Rourke and Williamson1999).

While this story is true in a macroeconomic sense, it is also incomplete. Almost all of the secondary literature cited in the first paragraph rests on a fundamental assumption of technological determinism. Financiers and financial institutions simply seem to react rationally to the exogenous appearance of technological innovations like the telegraph. Competition and contestation within financial institutions (let alone between them) for control over new technologies are considered negligible in the face of the aggregate efficiency gains new technology brought (Pardo-Guerra, Reference Pardo-Guerra2019, pp. 7–8). Recent research has challenged the empirical dimension of this approach directly by questioning the extent to which the telegraph actually did bring efficiency gains to financial markets (Philippon, Reference Philippon2015; Davies, Reference Davies2018; Bliss, Warachka, and Weidenmier, Reference Bliss, Warachka and Weidenmier2020). But there is also a more deceptive way in which this technologically deterministic approach has warped our understanding of the telegraph’s role in the history of finance. “The telegraph” in this literature appears as an abstraction – a generalized catchall that occludes the wide variety of different technological apparatuses, cables, circuits, and administrative practices which actually operationalized the telegraph within the financial system.

By viewing the telegraph not merely as a static technical object but as a financial infrastructure, a different understanding of the relationship between finance and the telegraph emerges. Thinking about the telegraph as an infrastructure highlights the relational and social nature of its functioning; that is, the technological gains in efficiency the telegraph brought were always shaped and distributed through social processes occurring within and between financial institutions. Put simply, the connections between financiers and financial markets that the telegraph enabled were unequal. Some of the world’s largest financial markets, like the London Stock Exchange (LSE), pushed for increased telegraphic connectivity in order to siphon business away from smaller exchanges globally. Many smaller exchanges saw increased telegraphic connection not as a harbinger of economic efficiency, but as a predatory threat, and thus resisted increased telegraphic connection. As the European powers jockeyed for imperial supremacy in the late 19th and early 20th centuries, geopolitics too played out within the technical details of the global telegraphic network. A delay or deficiency in one empire’s telegraphic system often led financial actors to route their orders through the telegraphic infrastructure and financial markets controlled by an imperial rival. The gains in efficiency that the telegraph brought inaugurated a conflict among financiers, financial institutions, and imperial states about who would obtain the benefits of that increased efficiency.

Questions about how the financial power and efficiency of the telegraph would be distributed were not only played out as a struggle between financial institutions and imperial states, but also occurred within a different set of relations: those between the technical object of the telegraph itself and the many human workers responsible for its operation and maintenance. Ostensibly mundane and seemingly apolitical technical questions about managing electrical current, repairing wires, and organizing message delivery systems took on increased importance as financiers, financial institutions, and imperial states looked for an edge over their competitors. Often, these technical questions revolved around managing the limitations of the telegraph’s infrastructural system rather than maximizing its efficiency. Many newer and faster telegraphs caused damage to the preexisting infrastructure of wires and cables or they contributed to an increased burden on the people tasked with transmitting and delivering messages, creating bottlenecks in the system. Small technical details within the telegraphic system very often had large financial and political ramifications.

Understanding the telegraph as a type of financial infrastructure allows us to see not only the aggregate gains in economic efficiency that new technology brought to financial markets, but how financiers, financial institutions, and states all struggled to incorporate new technology and reshape it for their own ends. In other words, how these institutions fought to capture the efficiency gains that telegraphic technology brought with it. Understanding the telegraph as a form of financial infrastructure allows us to move between different scales of analysis, connecting the technical challenges inherent in simply making the system work to the larger political-economic conflicts about who that system would work for. The rest of this chapter turns to one particular example of this phenomenon, particularly the LSE’s attempts to influence both the domestic and global telegraphic network in the late 19th and early 20th centuries.

2 London’s Telegraphic Dominance

Power differentials between core and periphery, and between imperial powers, have been a central concern in the history of telecommunications (Headrick, Reference Headrick1981; Bayly, Reference Bayly1996; Hills Reference Hills2010). Yet these power differentials are surprisingly absent within histories of the telegraph and finance. It is a striking omission, especially considering just how central financial business was to the profits of telegraph companies and how stridently financial institutions strove to reshape the global telecommunications infrastructure for their own advantage. As recent debates around high-frequency trading have revealed, when market information and orders are transmitted at faster rates, competition for small edges in transmission speed and minor reconfigurations of market structure become more important to market actors, not less (Easley et al., Reference Easley, Lopez de Prado, O’Hara and Zhang2021; MacKenzie, Reference MacKenzie2021). In other words, as the introduction of the telegraph sped up connections between all global financial markets, the competition between markets and firms for seemingly small edges in telegraphic transmission speed became ever more intense (Robertson, Reference Robertson2024). While the telegraph helped integrate markets on a long-run, macroeconomic scale, small differences in telegraphic connectivity had distributional consequences in determining which markets and firms profited from global financial integration. Speed was always relative.

London sat at the center of the global telecommunications network in the nineteenth century and exercised the most power over the network’s development (Wenzlhuemer, Reference Wenzlhuemer2013, pp. 271–273).¹ Britain’s geographical placement off the coast of Europe made it the natural meeting point for telegraphic cables passing between Europe, on the one hand, and North America, the Caribbean, and South America on the other. Advantages of geography were reinforced by the might of the British Empire. Imperial security concerns – and Britain’s raw ability to enforce them – helped create a network of direct telegraphic cables between Britain and the far-flung corners of its empire (Kennedy, Reference Kennedy1971; Headrick, Reference Headrick1991; Headrick and Griset, Reference Headrick and Griset2001). The British financial sector, especially the LSE, was another key factor in abetting London’s rise to telegraphic centrality. As the largest and most liquid financial market in the world, the high amount of daily business that was transacted via the telegraph on the LSE made it a profit center for many telecommunications companies who began to cater their services specifically for the use of Britain’s financiers.

Thanks in large part to the telegraph, the LSE’s reach was truly global (Cassis, Reference Cassis2006, pp. 98–100). As one observer from the New York Stock Exchange wrote in 1912: “the London Stock Exchange is the only really international market of the world. Its interests branch over all parts of our globe” (LSE, 1911, p. 353). Because of Britain’s deep pool of investors and the concentration of capital in London’s banks, many companies operating outside of Britain in the 19th and early 20th centuries dual-listed their securities in their local market and on the LSE (O’Sullivan, Reference O’Sullivan2015). The proliferation of dual-listed securities alongside the expansion of the telegraph network led to the explosion of specialized securities arbitrage. Communicating via the telegraph, arbitrageurs in London would work with local partners in markets throughout the world to eliminate the pricing discrepancies in dual-listed securities. London’s telegraphic centrality enabled extensive arbitrage operations, which in turn created a powerful network effect, as London could offer the most liquid markets in most securities combined with the closest prices, cementing the LSE’s place as “the centre to which the bulk of investment and speculative business naturally flows” (Armstrong, Reference Armstrong1939, p. 141).

The scale of the LSE’s business gave it a tremendous amount of power not only over the global financial system but also over the telegraph network. The exchange and its brokers were some of the first clients of new telegraph companies during the 1840s and 1850s (Baines, Reference Baines1898, p. 158; LSE, 1842–1852, p. 309). Even after the domestic telegraph service in Britain was nationalized and brought under control of the General Post Office (GPO) in 1870, the LSE continued to exercise considerable influence over how the telegraph was operated. Nationalization of the telegraph posed a set of difficult tradeoffs for the GPO to manage. While the GPO did not have to worry about turning enough of a profit to pay a dividend like a private company, individual departments did have a mandate to be economically “self-supporting” (Daunton, Reference Daunton1985, pp. 83–84). The GPO faced a tradeoff between supplying a neutral and fair telegraphic service to all users of the network at a higher cost for everyone or catering specifically to the needs of financiers in order to secure revenues that would subsidize cheaper use of the telegraph by members of the public. Throughout the 1870s and 1880s the GPO opted for the latter, allowing stock exchanges throughout Britain to have privatized telegraph offices and preferred access to public telegraph wires in order to drive up revenues from the financial system and subsidize the costs of telegraphs for the rest of the public.²

Preferred access to the telegraph network, and the advantages such access conferred, were not distributed equally even among financial institutions. The LSE benefitted disproportionately from the system of direct “stock wires” that was set up in the 1880s between its trading floor and those of the major provincial exchanges, including Liverpool, Manchester, Glasgow, Leeds, and Dublin. Telegraphic connection between London and the provincial stock exchanges quickly began to erode the uniqueness of provincial share lists (Campbell, Rogers, and Turner, Reference Campbell, Rogers and Turner2020). Shares were increasingly dual-listed on their local provincial exchange and on the LSE, which led to a robust arbitrage business among British exchanges. But due to the LSE’s massive advantage in liquidity over the provincial exchanges, the profits of arbitrage trading between London and the provinces were vastly unequal.³ One of the managers of the Liverpool Stock Exchange estimated that twenty times as much business was sent from Liverpool to London against what was sent from London back to Liverpool (Thomas, Reference Thomas1973, p. 89).

The LSE was even able to influence the development of the telegraphic network in more obviously unequal ways. During the late 1880s, a number of the provincial exchanges formed a plan to establish direct “stock wires” between each other in order to combat London’s monopolization of the arbitrage business. The plan was thwarted by the GPO and the LSE, as well as some of the major provincial exchanges themselves. The Dublin Stock Exchange, for instance, was already displeased at how much business they had lost from the telegraph and believed that establishing more telegraphic links of any kind would “take business away” from their exchange (Thomas, Reference Thomas1973, p. 103). The GPO, meanwhile, moved swiftly to protect the massive revenues it received from the LSE and subsequently passed a resolution to “discontinue direct (telegraphic) communications between all provincial offices that have not a direct wire to the London Stock Exchange” (Thomas, Reference Thomas1973, p. 103). Smaller exchanges, like Birmingham, Sheffield, Halifax, Huddersfield, and Bradford, which had pushed for a network of telegraph wires between provincial exchanges, were instead forced to compete amongst each other for direct wires to London, which continued to siphon off the majority of the arbitrage business for itself (BT, 1898–1927).

3 Global Telegraphic Influence and Its Challenges

Imperial competition between Britain, France, and Germany mounted in the 1890s and global telegraphic infrastructure was a crucial stage upon which this competition was played out. Germany and France were particularly frustrated by Britain’s global dominance of telegraphic infrastructure and began to lay their own telegraph cables to circumvent those controlled by Britain. The most impactful of these challenges to British telegraphic supremacy came from a German-laid transatlantic cable in 1900 that connected the German city of Emden to New York City via a waystation in the Azores. It was the first cable that allowed for transatlantic telegraphs to entirely bypass Britain and was so successful that another cable along the same route was laid just two years later (Headrick, Reference Headrick1991, pp. 105–107). In response to the telegraphic challenge from its imperial rivals, the British state began taking a firmer hand in managing its global telegraphic infrastructure (Kennedy, Reference Kennedy1971; Headrick and Griset, Reference Headrick and Griset2001, p. 552). This included the nationalization of the Submarine Telegraph Company in 1890 and the GPO’s subsequent management of the telegraph service between Britain and Europe.

London’s arbitrageurs had long been frustrated with the poor quality of the telegraph service that connected their exchange to the principal European exchanges. Delays and telegraphic traffic jams were frequent (Financial Times, 1891, p. 1). Since arbitrageurs took advantage of small price discrepancies in the same security that was traded on two or more different markets, even a slight delay in the transmission of a message could be the difference between a profitable or unprofitable transaction. Thus, slow transmission or delays in telegraphs to or from London led financial actors to route their orders to different exchanges on more efficient telegraph routes, depriving London’s arbitrageurs of business and the LSE of liquidity and close prices.⁴ For instance, the telegraphic connection between London and Italy was so slow that arbitrageurs in the key Italian bourses – namely Milan and Rome – often routed their orders to the Berlin Bourse first and only then on to London if an arbitrage opportunity still existed (Statist, 1896, p. 749). Other major cities and their financial markets – like Vienna and Budapest – still did not have a direct telegraph connection with London but they did have one with Berlin, and thus also routed most of their arbitrage business to the Berlin Bourse. Without liquid markets and close prices (provided in large part by arbitrageurs), the incentives to route orders to London would further erode and the network effects that had solidified the LSE’s financial dominance could be undone. Telegraphic dominance and financial dominance were closely intertwined.

Telegraphic transmission from Britain to the continent was so bad that even British financiers were beginning to circumvent British-controlled telegraph cables when routing their orders to Europe. In 1886, direct telegraph wires were laid that connected Liverpool to Le Havre in France and Hamburg in Germany, allowing Liverpool to send telegraphs directly to continental Europe without routing them through London first. However, almost all of the traffic from the North of England was also routed over these wires, creating constant telegraphic traffic jams and transmission delays. For “all intents and purposes,” a delegation of Liverpool financiers complained, they were not “in direct communication with (Continental Europe) at all” (BT, 1911–1921, File XVI). While arbitrageurs from Liverpool and elsewhere in the British provinces could route their orders to Europe through a relay in London, doing so meant that London arbitrageurs would almost always be a step ahead in closing arbitrage opportunities in European markets. Instead, arbitrageurs in the British provinces began to bypass London entirely by sending messages to New York via American- and British-controlled cables, from where they were relayed over the new, German-owned cable to continental Europe. Even though this route was physically longer, the time difference between New York and Liverpool meant that the transatlantic cables were relatively traffic-free during the European business day, so messages could be relayed and transmitted more quickly (BT, 1911–1921, File XVI). Delays and telegraphic traffic jams on the routes between London and financial markets in continental Europe caused more and more financiers to route their orders over German-controlled telegraphic infrastructure to German markets.

The decision to nationalize the Submarine Telegraph Company gave the LSE a unique moment of leverage over the GPO, and the exchange fought hard to win provisions that would improve its telegraphic connections to Europe. The LSE threatened to raise the rent on the GPO’s new offices significantly, from the £250 per year that the Submarine Telegraph Company had previously paid to £500 per year. However, if the GPO established direct telegraphic communication between London’s trading floor and those of the key European exchanges like Paris, Amsterdam, Brussels, Berlin, Frankfurt, and Hamburg, the exchange was willing to waive rent altogether (BT, 1889–1946, File I). Up until that point, telegraphs were handed in at offices located in the stock exchange but were sent via pneumatic tube to the GPO’s Central Telegraphic Office a few blocks away, from where they were transmitted to the continent. With few exceptions, financial telegraphs arriving in Europe would similarly have to be transmitted from a central office and then relayed to the local stock exchange. Direct “bourse-to-bourse” service attempted to cut out these middle relays within the telegraph network. The GPO, eager to save money on rent and hopeful that the new stock exchange wires would increase business, quickly agreed to the arrangement. But the LSE and the GPO would both come to find that influencing the global telegraphic network would prove much harder than influencing the domestic one.

Establishing direct telegraphic communication between financial markets was a far trickier proposition in practice than it was in theory. There were two major problems. The first was technical. How would the GPO actually implement the direct telegraphic connection between the LSE and its European counterparties? The second problem was administrative. How could the GPO convince European telegraph services (most of which were imperial rivals with Britain) to prioritize the transmission and delivery of financial telegraphs?

The technical problem of creating direct telegraphic connections between stock exchanges was exacerbated by Britain’s position off the coast of Europe. Any new cables laid between London and Europe had to be undersea cables, which were significantly more expensive and technically more difficult to lay than overland cables. This was financially unfeasible for the GPO. Instead they explored two different options: diverting existing telegraph wires exclusively to stock exchange business and introducing new, more advanced telegraph apparatuses that would enable a higher volume of messages to be sent over existing wires.

Diverting wires for exclusive stock-exchange business was both technically and administratively challenging. It also highlighted the stark differences that existed in telegraphic connectivity between different cities and financial markets in the late nineteenth century. On the one hand, cities like Paris had extensive telegraphic connections with London. By 1889, there were 11 telegraphic wires connecting the two cities and, at the peak of the business day, over 280 telegraph messages per hour were sent between the LSE and the Paris Bourse. “If there is pressure of business of any particular kind it is spread over the whole eleven wires,” a GPO official noted, but if five wires were diverted for exclusive bourse-to-bourse service, “a pressure of Stock Exchange business could only be met by the use of those 5 wires and the remaining six would not be available” (BT, 1889–1946, File III). Alternatively, many of the smaller exchanges in Europe did not have enough traffic to merit an exclusive wire. Antwerp, Brussels, and Hamburg all maxed out at between fifty and sixty stock exchange telegraph messages per day (BT, 1889–1946, File II). It was only the stock exchanges with a medium volume of messages where the GPO believed an exclusive wire would be a benefit. Amsterdam, Berlin, and Frankfurt all averaged between 150 and 250 stock exchange telegraph messages per day. Each city had more than enough telegraph wires running into such that diverting one wire exclusively for stock exchange business would not affect the telegraphic capacity of the rest of the system. There was no way to stand up a true system of bourse-to-bourse telegraphy across such an uneven telegraphic landscape. By the end of the 1890s, the GPO had quietly given up on the project of diverting wires to the direct bourse-to-bourse service because it was an inefficient solution to the problem.⁵

The complaints from the LSE and its arbitrageurs, however, did not abate (LSE, 1910–1913, pp. 143, 169). The GPO, in turn, searched for new solutions to the problem, namely the introduction of more advanced telegraph machines. If higher volumes of traffic could be processed on fewer telegraphic wires, more wires could be exclusively committed to stock exchange business without system-wide slowdowns. Two new telegraphic apparatuses invented by David Hughes and Émile Baudot offered a possible solution. Hughes and Baudot telegraphs specialized in duplex and quadruplex telegraphy, respectively. As the terms suggest, duplex and quadruplex telegraphy enabled two and four messages to be sent along the same wire in different directions. This was a major improvement upon previous systems of telegraphy that operated on a simplex system, which could only send one message in one direction over a wire at a time. Simplex systems were easily overwhelmed by even a moderate amount of traffic, and led to particularly financial forms of manipulation. Liverpool brokers complained that when they attempted to telegraph Germany via their direct, simplex wires, German firms could simply “block the wires” by flooding them with telegraph messages in the opposite direction (BT, 1911–1921, File XVI). Not only could Hughes and Baudot telegraphs handle a higher volume of traffic, but they also automatically printed their messages out, rather than waiting for a human operator to transcribe them. All told, the average words per minute increased from twelve to fifteen on Morse telegraphs, to thirty-one on Hughes telegraphs, to thirty-five to forty on Baudot telegraphs (Butrica, Reference Butrica1986, pp. 78–83, 97–107).⁶

Even though multiplex telegraphy brought significant gains in speed and efficiency, its infrastructure was also more fragile. Multiplex telegraphy worked by splitting the electrical signal it sent over the wire, which made it especially hard to manage on long wires (where the signal would weaken) and undersea wires (where the saltwater created increased electrical interference). Moreover, wires within the same cable that were operated with a quadruplex system frequently created electrical interference with other wires that were operated on a simplex or duplex service (BT, 1911–1921, File VII). The technical issues involved were so thorny and persistent that even as late as 1935, only three out of the thirteen critical and busy transatlantic cables utilized any form of multiplex telegraphy at all (Federal Communications Commission, 1936, p. 44). Many of the gains in efficiency achieved by more advanced telegraph apparatuses were limited or complicated by the technical fragility of the infrastructure on which they depended.

It is here that labor and administration – the other key problem the GPO faced – came into play. Not only were the more advanced, multiplex telegraphs dependent on a fragile infrastructure, but they were also technically much harder to operate than earlier telegraphs. Operators needed to have a knowledge of electrical engineering in order to troubleshoot problems and keep the telegraphs in synchrony with each other, on top of their usual duties of rapidly transmitting messages (Butrica, Reference Butrica1986, pp. 115–116; Hindmarch-Watson, Reference Hindmarch-Watson2020, p. 65). In Britain, expansion of Baudot telegraphs went slowly in part because the GPO was careful to make sure each operator had enough experience handling the apparatus before they were assigned to operate an entire circuit on their own (BT, 1911–1921, File VII). Similar problems regarding the training of operators plagued the European service as well. Even though the French telegraphic system had been early adopters of Baudot telegraphs – putting over 100 Baudot machines into use by 1891 – French telegraph operators and engineers struggled to implement duplex and multiplex working throughout their service, preferring to use the more advanced machines of Baudot to simply accelerate their existing simplex telegraphic services (Butrica, Reference Butrica1986, pp. 99, 112–113). When the GPO began attempting to expand duplex and multiplex telegraph service throughout its European circuits, British engineers did not hide their frustration with their French counterparts. The “French administration,” one engineer complained, was known to be “not favorably disposed to duplex working as a general principle” (BT, 1911–1921, File VII). The difficulty of operating and maintaining advanced telegraph apparatuses created new bottlenecks in training a sufficient number of operators for the telegraphic workforce.

Frustration with European telegraph operators was not the only administrative hurdle the GPO faced. In Britain, telegraph messages were delivered from the transmission offices to clients via a small army of telegraph messenger boys. At its peak, the GPO employed 3,200 messenger boys in 1897 and trained them with a military-style rigor (Hindmarch-Watson, Reference Hindmarch-Watson2020; Joyce, Reference Joyce2013). Many arbitrage firms employed their own private messenger boys as well. Life as a messenger boy in an arbitrage firm entailed less militaristic discipline but placed a premium on physical speed. As one stockbroker told a messenger boy in his firm, “in the City, as young man you never walk, but always run” (Janes, Reference Janes1963, p. 47). The same was not true in continental Europe. British arbitrageurs complained that while “sharp boys” delivered telegrams in Britain, European telegrams were delivered by “army pensioners and other middle aged or old men, sometimes women who are naturally slower than boys” (BT, 1911–1921, File XVI). The European delivery systems had other drawbacks as well. For instance, telegraph messengers in Paris were instructed to “wait until there are a number (of messages) for the same district,” rather than delivering telegraphs individually as they arrived (BT, 1911–1921, File XVI). In order to avoid the delays that occurred when messages were transmitted via the different national telegraphic systems, British arbitrageurs increasingly routed their orders to Europe via private American telegraph companies, like Western Union. Unlike the GPO, private American telegraph companies could use their own messenger boys in European offices and were therefore able to deliver messages “with the same expedition” that customers had come to expect in Britain (BT, 1911–1921, File IV). Creating a direct, bourse-to-bourse system of telegraphy was beset with technical and administrative challenges that bedeviled the GPO for years.

Gradually, the problems of implementing the bourse-to-bourse system began to give way. By 1911, the GPO had successfully introduced multiplex telegraphy to a number of its domestic telegraph circuits. Armed with an adequate number of highly trained operators, and some key technical innovations – like the adoption of anti-induction devices which helped ease problems of electrical interference between wires in the same cable – the GPO began to push for the introduction of Baudot multiplex working on the key European circuits that connected London to Paris and Berlin (BT, 1911–1921, File VII). Buoyed by the success of Baudot working with Paris and Berlin, the GPO expanded Baudot usage to the Amsterdam circuits and the long-troublesome Milanese circuit by the end of 1912. They also laid further plans to aggressively increase Baudot working to the rest of the Netherlands, as well as Brussels, Hamburg, Frankfurt, Emden, and even to Budapest and Vienna, which had previously not had direct channels to London at all (Table 10.1).

The expansion of Baudot to Europe finally gave the GPO enough open telegraphic channels at their disposal to dedicate telegraph circuits exclusively to the “bourse-to-bourse” service. Rather than creating all new telegraph circuits and wires, the extra telegraphic capacity allowed the GPO to reallocate the circuits that serviced the bourses of Paris, Berlin, Frankfurt, Hamburg, Amsterdam, and Brussels to an office in the LSE for exclusive financial work (Table 10.2). For instance, the two Baudot circuits, which provided eight telegraphic channels between London and the Amsterdam central offices, enabled the GPO to transfer two Hughes circuits to exclusively service financial telegrams between the LSE and Amsterdam Bourse. The Paris and Berlin bourses were the only exchanges that received their own dedicated Baudot apparatus in addition to supplementary circuits operated by Hughes machines. Some administrative issues surrounding the delivery of messages remained. The telegraph office in the Paris Bourse handled both financial and ordinary telegraphs and it was not clear how to administratively disaggregate the two types of messages to ensure speedy delivery (BT, 1889–1946, File III). In London, the GPO still worried that splitting the operation of key circuits between the Central Telegraph Office and the LSE would necessitate an “amount of staff, considerably in excess of what would meet the requirement if the whole of the circuits were concentrated in one office” (BT, 1911–1921, File VII). But these were ultimately minor issues since “the disadvantages of working certain of the continental circuits from a point in London other than the Central Telegraph Office cannot continue to be regarded as outweighing the advantages of establishing direct communication between the Stock Exchange in London and the Continental Bourses” (BT, 1911–1921, File VII). In August 1913 the GPO signed off on a plan to establish a bourse-to-bourse system of telegraphy between the LSE and its European counterparts.

But the best-laid plans – and telegraph wires – sometimes go awry. When World War I broke out in August 1914 stock exchanges closed, international communication between warring nations was severed, and arbitrage trading became “practically impossible” (The Times, 1914, p. 17). The GPO and LSE abandoned their plans for implementing the bourse-to-bourse service. It would take another seven years before bourse-to-bourse telegraphy from the LSE was actually established, in 1921. Even then, the system was far more limited in practice than had been imagined before the war. Only the Paris and Amsterdam bourses were afforded direct telegraphic channels to and from the LSE and, notably, no German exchanges were included. Bourse-to-bourse service was also limited to the busiest hours of the business day (BT, 1911–1921, Files III, IV). Thirty-one years after the idea had first been proposed, a bourse-to-bourse telegraph service was finally in operation. But thanks to persistent problems inherent in managing complex technical infrastructure, sourcing enough staff to operate all the components of the system, and the shifting, explosive consequences of geopolitics, the bourse-to-bourse telegraph system was only ever implemented on a limited scale.

4 Conclusion

Focusing on the particularities of telegraphic infrastructure, rather than just the telegraph in the abstract, challenges many of our assumptions about the relationship between technology and financial markets in economic history. By paying closer attention to the operational details of telegraphic infrastructure, we can begin to move away from big bang accounts that emphasize a single moment of exogenous, technological innovation. Once the telegraph was introduced to financial markets and institutions, its benefits became relative. Telegraphic speed and capacity, as well as the economic benefits such speed and capacity bestowed, were always unequally distributed. Firms, exchanges, and states all had different imperatives when choosing how to implement and use the telegraph. Increased telegraphic connectivity was not always seen as an unalloyed economic good. The competing perspectives of different financial actors about how best to configure the telegraphic system had material consequences for how global telegraphic infrastructure was built and operated. Macro-level conflicts between different states and markets were often played out within the seemingly small and technical space of telegraphic infrastructure. This meant that the power of these states and financial institutions often rested on the infrastructural limits of the telegraphic system and on the workers whose job was to make that system run. Understanding the telegraph as a financial infrastructure, then, allows us to see how these seemingly small battles around technical details had large ramifications for the political economy of the financial system as a whole.