The term “service sector” sounds vaguely anachronistic. The twenty-first century, often referred to as the post-industrial age, is one of the unmistakable dominance of service industries. The great economic historian R.M. Hartwell (1971) once lamented that the service sector was the “neglected variable” in the economic history of Britain at this time, and that our knowledge of what transpired with this variable during the Industrial Revolution was rather modest compared to the huge amounts written on manufacturing and agriculture.
The period 1700–1850 was hardly an age in which all production took place in the tangible sectors of agriculture and manufacturing. The service sector was already quite large in 1700 (estimated at 34 percent of national product in 1688, see Deane and Cole, 1969, p. 156) and kept expanding, though exact numbers are hard to take seriously given the ambiguities of the data and of the definition of what the sector includes. Lee (1986, pp. 9–11, 98–100) has computed that, with the exception of the years 1815–40, the contribution of services to aggregate growth was at least as large as that of manufacturing. The same is true for employment growth. To some extent, dividing the labor force into such sectors quantitatively ignores the difficult problem that many workers were doing part-time work in different sectors. The sectoral distribution of workers assumes a level of specialization that had not yet been attained. Yet it deserves attention despite the roughness of the estimates. For one thing, perhaps the most common single occupation of the British working population in the mid-nineteenth century was that of domestic servant, a profession that, save for a small minority, has passed away from the industrialized world. In 1841, however, there still were 984,357 “domestic servants” in Britain out of a total labor force of 16 million (Great Britain, 1844b, p. 283), the largest reported occupation by the 1841 census, slightly ahead of “labourers, agricultural” (960,382) and almost five times larger than “cotton manufacturers, all branches” (213,944). Transportation, too, was a major employer in 1841, with 33,867 adults employed as carters and wagoners and 45,915 sailors on shore (the ones at sea were missed by the census). Other service sectors that were still small compared to the dimensions they were to attain in our own time were education, finance, government, and health. Even if these sectors were still rather modest in size compared to, say, agriculture, they contributed disproportionately to the economy, much as a lubricant to a well-functioning engine. Above all, however, Britain bore out the dismissive remark attributed to Napoleon that Britain was a “nation of shopkeepers.” In fact Adam Smith had coined the phrase and Napoleon may have been familiar with it. It was a commercial nation, one in which trade, shipping, and transportation played roles of great importance. The basic Enlightenment concept of progress and enrichment through the systematic accumulation of useful knowledge and its rational application can be seen to play a major role in the service sectors as well, although it was not as palpable and concrete as it was in the production of textiles or hardware. Yet there was no automatic transition from a commercial nation to sustained technological progress and industrialization. What was the service sector’s function in the creation of modern growth?
* * *
Gregory King ([1688], 1936) estimated that out of a population of 5.5 million, 180,000 people were shopkeepers or their families, and another 54,000 people were “Merchants & Traders by the Sea.” Altogether, this would point to a total of 4.2 percent of the population in “commercial” activities. There were also 240,000 people in occupations he called “artizans & handycrafts”—and many of those must have spent some time selling the goods they made. Modern research (Lindert, 1980, pp. 702–03) has established that in 1688, about 9.6 percent of all males had occupations that can be classified as “commercial.” For 1700, the corresponding number is 8.3 percent. In 1755 commercial occupations rise to 13 percent, but afterward the numbers seem to stabilize and even to decline slightly relative to the population. The census of 1811 reports only 5.9 percent of the population as part of the commercial classes, but the computations in Lindert and Williamson, 1982, suggest a total of 205,800 merchants and shopkeepers in 1801, which would be about 10 percent of households. The later and more reliable censuses of 1841 and 1851 make such computations cumbersome, because of their odd and inconsistent ways of classifying workers. Among the adult males surveyed in the 1841 census, we can add all occupations that were clearly mercantile, and find that 305,625 or 16.7 percent of employed males over 20 declared occupations such as “grocer,” “dealer,” and “broker.” For adult women the number is 3.55 percent (largely because 45 percent of all women over age 20 were employed as “servants”). These numbers seriously understate the real number, because this estimate excludes a substantial number of people who classified themselves as having both a manufacturing and a commercial occupation, such as the 114 “pen-makers and dealers.” Independent artisans sold their own wares, and some of those occupations were of course numerous (bakers and shoemakers alone accounted for 243,000 adult males). Long before the Industrial Revolution, Britain was a commercialized nation which relied on markets, where supply met demand, money changed hands, credit was widely used, and competition was often fierce.
All the same, commerce changed considerably in the period 1700–1850. Urbanization meant, inevitably, a growth in the number of customers who raised little or no food themselves and were entirely dependent on what they could purchase. But these were also years of growing occupational specialization, in which occupations began to mean more what they mean today: some people made things, others maintained and repaired them, still others sold them, and these functions were slowly becoming more separate although the process was far from complete by 1850. Foreign visitors were astonished by British stores. In 1706 the Frenchman de Souligné (probably a pseudonym) thought that ancient Rome had not “the Fourth part of Shops, Arts and Handicrafts as we have in London” (1706, pp. 100–01). In London “the magnificence of the shops and warehouses, which often extend without interruption the length of an English mile are peculiarly striking,” wrote the German visitor J. von Archenholz ([1785], 1797, p. 145) in around 1780. Even within commerce, growing specialization began to occur, with wholesalers becoming distinct from retailers in some sectors. With the integration of markets and the rise of a “national market” in some commodities, more and more people ended up purchasing from strangers. If local carpenters and brushmakers had supplied most of the needs of the average family in the time of Elizabeth I, in the eighteenth century regional or national markets in many goods began to emerge. By the late eighteenth century Frederick Eden found that while in Elizabethan times husbandsmen still wore coarse linen made at home, this held by his time only for the northern counties and Scotland. In the Midlands and southern counties the laborers purchased most if not all of their clothes from the shopkeeper whereas in the north there were still many respectable persons who never “wore a bought pair of stockings, coat, nor waistcoat in their lives” (Eden, 1797, Vol. 1, pp. 120–21, 554–55). The “moral economy,” in which people traded primarily with people they knew personally, slowly gave way to a more sophisticated, complex, impersonal system. Many contemporaries decried this loss of innocence, suspecting middlemen of profiteering, and voiced their suspicion of the laws of supply and demand. Ideas now taught in every business school such as “marketing” and “advertising” slowly picked up. Special clothing halls to sell woolen products were opened, and traveling salespeople with their samples and drawings charged into expanding markets. Consumers were constantly told that the world of goods was a good world, the only world worth living for. In the eighteenth century trading cards issued by merchants became an effective means of bringing one’s goods to the attention of customers as well as a gentle reminder to customers who owed a balance (Hubbard, 2009).
The notion of marketing was pushed furthest by the famed potter Josiah Wedgwood, whose appeals to snobbery and to the nobility-envy of the merchant and middle classes were an early example of what some might think of as consumer manipulation. Wedgwood’s marketing strategies included a brazen display of goods targeted at the high and the mighty, to be imitated by the would-have-beens and even by the never-were. It is easy to exaggerate the representativeness of Wedgwood: he was in many ways a highly unusual individual, an entrepreneur of rare imagination and audacity, and few could measure up to him. But he did set an example, and aggressive marketing strategies can be discerned in other industries such as printing, cutlery, clocks, high-end textiles, and household implements, to say nothing of medical doctors and pharmacists selling miracle drugs. Advertisers shamelessly dropped the names of peers and royalty, whether honestly or not. Matthew Boulton mentioned the King’s architect, clockmaker, cutler, and physician in his correspondence and clearly cultivated such relationships (Robinson, 1963, p. 50). Marketing became an industry. A well-known example is that of the razor salesman George Packwood whose advertising campaign in the mid-1790s was a “remorseless attempt to imprint the brand name on the public memory” (McKendrick, 1982, p. 148). Advertising revenues became an ever more important source of income for newspapers. Provincial newspapers teemed with advertisements, particularly appealing to the snobs by alerting them to “metropolitan tastes”—that of London, above all. Advertising may seem far removed from ideas of Enlightenment, but free competition, ingenuity, and the dissemination of information, all core values of the Industrial Enlightenment, were what advertising were all about. So, of course, were dissimulation and consumer manipulation.
With declining transport costs in the late eighteenth century, more of London’s consumer culture filtered down even to small provincial towns, who aped it wherever possible. Coffee-houses, opera companies, and lending libraries emerged all over provincial Britain. Theaters with names such as Drury Lane sprouted up in small cities. A visitor to Halifax in 1781 was astounded by the quality of the local bookseller. London may not have been an industrial and technological leader during the the Industrial Revolution, but it still set the tone of demand patterns. The wool industry, especially, increasingly catered to fashion, understood it, and manipulated it. The cotton industry’s great success was in part based on the ability of cotton fabrics to absorb printed patterns and colors, which made them attractive to consumers, especially after 1783 when Thomas Bell perfected the technique of printing patterns on cotton cloth using copper cylinders. Dr Johnson sighed that “Promise, large promise, is the soul of advertisement. I remember a Wash-ball that had a quality truly wonderful; it gave an exquisite edge to the razor … The vender of the Beautifying Fluid sells a Lotion that repels pimples, washes away freckles, smooths the skin, and plumps the flesh; and yet, with a generous abhorrence of ostentation, confesses, that it will not restore the bloom of fifteen to a Lady of fifty … The trade of advertising is now so near perfection that it is not easy to propose any improvement” ([1759], 1800, Vol. 1, pp. 135–37). This may seem somewhat comical in the twenty-first century, yet clearly indicates that by this time commercial advertising was a common sight.
Retailers found new ways to reach customers. Samples were sent around with traveling salesmen, and displayed in country inns. Warehouses and display rooms were set up in places like Bolton (outside Manchester), although eventually much of this activity moved to the city itself. Some manufacturers opened permanent showrooms (e.g., the cutlery works of Rodgers & Sons in Sheffield). In Halifax, the cloth hall was opened in 1775, and had 300 rooms in which trading and the exchange of information could take place. Firms also increasingly employed overseas agents to help them market products abroad. Some manufacturers decided to do their own marketing, while others relied on wholesalers. Wholesale trade emerged slowly over the eighteenth century, but by 1850, the separation between it and retailing was already quite advanced. The result of the gradual expansion of marketing, new and better goods, more choice and lower prices was that in eighteenth-century Britain consumerism—some would say materialism—became a significant social force. It affected the choice every consumer had between market-produced goods and home-made ones, and helped shift preferences toward purchased goods. Yet in order to buy goods, households needed cash, that is, to work more outside the home or sell the products they made rather than consume them. This gave rise to what has become known as the “Industrious Revolution”—a growth in the market-oriented labor supply driven by a desire for consumer goods (De Vries, 1993, 1994, 2008).
* * *
Of the many “revolutions” that were supposed to have taken place in Britain between 1700 and 1850, the transportation revolution occupies a pivotal role, in that it affected all other sectors in subtle but pervasive ways, and was itself subject to the institutional and technological advances that changed the British economy. Transportation is a technology, but it also needs to be organized, coordinated, and financed in ways that are special. Transportation usually requires substantial overhead investment as well as a physical layout as a network, in which complementary and rival lines are often very close to one another. The history of modern transportation was punctuated by the development of the railroad in the late 1820s, and some historians have viewed the railroad as the primary driving force of modernization and growth. There is more and less to this story than is thought: transportation development included much more than railroads, which, however, may not quite have had the dramatic impact believed by some in the period covered here. Much like the argument I made for agricultural and industrial steam power, the real factor that transformed the economy was the general drive for progress and not just its various manifestations. Long before the emergence of the steam locomotive, transport costs were declining. Thus, better-built roads and improved coaches sharply reduced internal travel time in the eighteenth century: the coach from London to Edinburgh still took 10–12 days in the mid-1750s, whereas in 1836 (just before being replaced by a railroad) it could cover the distance in 45½ hours.
Assessing the exact impact of transportation improvements on economic development is far from easy. The movement of goods and mobile factors of production within a country permits regional specialization, which, as economists never tire of explaining, makes all regions better off. Better transportation will lead to better allocations of capital and labor. Following the development of the transport network, British labor became more mobile. Workers could travel around looking for jobs, and in some cases work for periods in remote places without arduous and long trudges across the country. Even capital seems to have been sensitive to these developments and became more mobile in the eighteenth century (Buchinsky and Polak, 1993). Adam Smith’s celebrated division of labor depended on tolerably low transport costs in the market: his idea of “the extent of the market” must be defined in the context of transportation capabilities. But there were more subtle and indirect effects as well: in much of the eighteenth century, Britain’s regional diversity was still quite pronounced, and ideology, politics, and culture expressed local values and interests as much as national or universal ones (Langton, 1984). It is this kind of cultural heterogeneity that better transportation and communications tend to erase, though diversity is never quite eliminated. Indeed, specialization may have enhanced regional differences, not just between manufacturing and farming regions but also between, say, the textiles of Lancashire and the hardware industries of the Black Country.
Better transportation weakened and possibly eliminated local monopolies and forced producers and merchants to compete with one another, a process that enhanced efficiency and speeded up the diffusion of new techniques. By unifying large markets, good transport tended to encourage the creation of standardized products and through it mass production, and encouraged investment in marketing and management (Szostak, 1991). A good railroad is the mortal enemy of the monopsonistic employer in the “one-company-town,” who could exploit his employees. More subtly and harder to observe, better transportation meant that ideas and knowledge could flow more easily across space and thus that they affected access costs. With the completion of the turnpikes, Boulton and Watt were able to travel from Birmingham to Paris in 1786 in the astonishingly short time of six days (Jones, 2008, p. 27). Ideas and information were carried by books, magazines, letters, and people, which could only move at the speed at which physical objects could move, a constraint that is easily forgotten in the age of the internet. As John R. Harris (1992) has argued, much of the tacit, crafts-based knowledge in the eighteenth century spread through the movement of skilled workers from one area to another, and “industrial espionage” remained an important part of the technology of access to knowledge. Specialists such as the consulting engineers needed for the maintenance of early steam engines and other machinery or coal viewers, the experts who assisted in the construction of deeper and more elaborate mines, could travel from site to site. Itinerant lecturers moved about in Britain and spread knowledge and techniques. Consultants and experts traveled about and helped diffuse technology. The roads between Cornwall and the southern Midlands were well traveled by Watt’s employees. Books, magazines, pamphlets, and letters spread more easily and quickly. Technology improved faster when inventors and mechanics could have good access to techniques used elsewhere, could compare notes, and could combine different and disparate ideas into new forms. To do this, they needed good access—and that was what better transportation provided.
Before the coming of the railroad, transport took three forms: road transport, ocean transport, and internal waterways. All three modes had their own technological and organizational problems that needed to be solved, and the period in question saw a plethora of solutions. None of those were perhaps as dramatic and as spectacular as the railroad, but economic effects are sometimes most penetrating and pervasive when the changes are least visible on the surface. In the case of transportation, at least, the changes were visible and have been documented by historians.
Road transport in Britain improved a great deal in the century before the railroad. Like every other economic sector between 1700 and 1850, progress came not from a single spectacular breakthrough but from a variety of sources. Technology improved at a multitude of levels: roads were better constructed, and stronger animals hauled better-made carriages. But improved institutions and organizations were equally important. Firms running passenger and freight services got larger and more efficient, through learning-by-doing or economies of scale (Gerhold, 1996a, p. 502). The most important development, however, arose from the evolution of the turnpike trust. Road-building and maintenance had traditionally been the responsibility of local authorities (parishes). The dilemma was that the benefits of free-access roads accrued to a large extent to non-residents in transit or their customers, and unless there was a way to collect money from these customers, there would be no incentive for local authorities to spend resources on roads. The coordination problem here was quite obvious: when a road led through many parishes, each parish was in charge only of the segment running through it. There was little point in the local authorities improving it beyond what the neighboring parish would do. Thus the mean quality of roads would gravitate to the level of quality consistent with the poorest or stingiest parish on the way, what is known today as a “race to the bottom.”
Turnpike trusts had emerged in the seventeenth century. A good example of institutional innovation, these trusts were established by Act of Parliament. Such Acts allowed local authorities to charge tolls from users in exchange for maintaining the quality of the roads. The tolls were regulated by Parliament, so this was hardly a free market, but it was a vast improvement on the chaotic system and poor-quality roads that the parishes had created. The first such trust was passed by Parliament in 1663, but in the seventeenth century these were largely confined to the London area. The decades 1750–70 witnessed the peak of the “turnpike boom” in which much of the rest of the country was turned into turnpikes. By the mid-1830s, about 22,000 miles of roads in Britain (about 17 percent of the entire road network, but including most of the important roads leading to big cities, including London) had been “turnpiked” and roads had improved significantly (Bogart, 2005a, 2005b). Using the best data available, Bogart has shown that the trusts reduced transport costs while improving quality, and that the effect of turnpike trusts came on top of the technological changes in road transport. The social savings methodology he employs indicates that around 1820, on the eve of the railroad age, turnpike trusts added around 1 percent to national income through lower transport costs, not counting the indirect and harder to measure spillover effects discussed earlier (Bogart, 2005a, p. 501). Furthermore, by comparing the roads that were turnpiked with those that remained under the traditional parish management (and adjusting for inevitable selection bias), Bogart (2005b) shows that turnpike trusts actually increased investment in road improvement and did not just replace funds that would have come from somewhere else. The quantitative evidence amassed by Bogart is complemented by evidence from firm records: turnpike roads were used to transport cotton goods in the 1780s and 1790s because they were faster and more reliable even if canals were less expensive (Freeman, 1980).
The turnpike movement coincided chronologically more or less with the Industrial Revolution. Its significance for an understanding of the era is that it cannot have been caused by the developments in cotton, steam, and iron, which were still in the future, nor was its impact on the economy large enough to account by itself for the technological developments. The only conclusion one can draw is that both were driven by a deeper phenomenon, which involved the ability to apply useful knowledge to practical problems as well as the political willingness to reform antiquated inefficient institutions to make the economy work more smoothly. The capability of British institutions to reinvent themselves and make the economy advance without bloodshed and political upheaval reform constituted an essential component of the success of the Enlightenment program in that country.
Institutional progress, even more than technological progress, was never linear and direct. As we have seen, the attempt to advance on many fronts ran into stubborn resistance in the form of “messy” technical problems that could not be readily solved with the knowledge base available. Even when they were solved, however, ex post concepts of progress could be misleading. Road travel and canals provide an example of how the relentless quest for improvements sometimes led to progress on fronts that eventually turned out to be a dead end. Stagecoach traffic through England increased rapidly during the eighteenth century. The number of public coaches leaving London each week (in the summer) was 465, of which 394 traveled less than 60 miles from the city. By 1783, this number had increased to 5,805, of which 3,735 traveled less than 60 miles (Austen, 1981, p. 26). The average speed of these coaches was slow, less than 4 miles per hour in the early 1760s, but by the early 1830s this had increased to about 8 miles per hour (Jackman, [1916], 1962, pp. 683–701; Bogart, 2005a, p. 484). The costs of travel did not fall much over the long haul, but costs were not everything in transportation: frequency, reliability, speed, and comfort all increased dramatically in the century before the first railroad was built. Stage-coaching, despite the costs and the discomforts, was very much the mechanism that provided Britons with the mobility on which so much depended, and one scholar, writing in about the 1830s, has remarked that in the industrializing areas the coaches provided the kind of communication that broke down barriers and gave these areas greater cohesion and unity than ever before. The idea of regular, reliable, and frequent passenger transport between towns was due to the stagecoach, not the railway (Dickinson, 1959, pp. 10–11). When the railroads came, passengers were ready for it, but the turnpikes lost their importance.
The age of Enlightenment was the age of improving communications and declining access costs. Much information was communicated through personal letters, and what a letter-writing age needed was a good postal service. In 1683, William Dockwra, a London armorer, set up London’s Penny Post. In an age of rent-seeking, however, natural monopolies were confused with revenue-generating government enterprises, and the Crown revoked his patent and took over the enterprise. During the eighteenth century mail services gradually expanded through byway posts (cutting out the need to go through London). These byway posts were the brainchild of a Cornwall postal employee, Ralph Allen, one of the unsung successful entrepreneurs of the first half of the eighteenth century, who bought the rights to all byway post in Britain. By the time of Allen’s death in 1764, most of England and Wales received mail daily (Headrick, 2000, p. 187).
The Post Office became a factor in land transportation with the introduction of mail coaches in 1784. The brainchild of John Palmer, the mail coach adopted an innovative design made by carriage maker John Besant in 1795 that used the famed “mail axle” that prevented the wheels from coming off accidentally. Palmer was a rather obsessive Bath theater owner, who became frustrated by the slowness of the mail, and single-handedly cowed and bullied the Pitt government into reforming the mail services. The mailcoaches carried passengers as well as mail, and although the share of passengers they carried remained small, the competition did the industry a lot of good, especially in long-distance travel. The process was completed in the mid-1780s. Mail coaches were a considerable improvement, with the emphasis on promptness, punctuality, and speed, traveling overnight and stopping only to change horses. They were absolved from turnpike tolls, and received priority over other vehicles. The postal system was wholly reformed, and Palmer was recognized as one of the most distinguished improvers of his age, and awarded a grant of £50,000 in 1813. In 1840 Rowland Hill established the national Penny Post, which became the standard of the efficient and accessible postal system that signifies the importance that Victorian society placed on good communications.
Better road-building technology added to these improvements. The period of the Industrial Revolution was famous for progress in the way roads were built. The “trio” of great road engineers consisted of John Metcalfe (1717–1810), most famous for his technique for draining rainwater on both sides of the road in ditches; Thomas Telford (1757–1834), who used uniformly sized stones for the foundation of his roads and small broken stones on the top, which got harder as horses’ hooves and carriage wheels compacted them solid; and John Loudon MacAdam (1756–1836) whose idea of building slightly convex roads earned him immortality as “Macadamization” (covering roads with layers of broken stones) spread far beyond Britain. Gentler gradients, a central component of road improvement, helped reduce horse exertion. Surface durability and the ease of drainage, however, were only two technical aspects of these improvements. Other forms of progress were equally important. One example of the happy marriage between ever better informed and more sophisticated engineers and the transportation sector was the construction of more advanced bridges. Of these, the most remarkable was Telford’s magnificent suspension bridge over the Menai Straits in Wales, completed in 1826.
With better roads came better-designed and lighter carriages, stronger and more efficient horses, and greater speed (Gerhold, 1996a). An index of productivity in road transport (measured inversely through costs), provided in table 10.1, shows that the efficiency of road transport on two selected and perhaps not typical routes (Leeds to London freight carrier, London to Exeter coaches) more or less tripled between 1700 and about 1830, with the greatest acceleration happening around the middle of the eighteenth century. Given that the truly discontinuous breakthroughs in road technology such as asphalt, pneumatic tires, and the internal combustion engine were still far in the future, this was a truly remarkable achievement. Improvements in internal transport were an important part of the continuous spread of knowledge and ideas essential to an enlightened economy. But beyond that, commodities needed to be moved around if economic efficiency was to be attained through competition and specialization. The improvements in roads were essential to an industrial town like Birmingham, which had neither a harbor nor a good river nearby. Its population tripled between 1700 and 1750, and tripled again between 1750 and 1800, making it the third largest town in England after London and Bristol in 1775. It success as the hardware capital of Britain depended on its ability to ship its buttons and toys through the entire kingdom.
Table 10.1: Productivity increase in road transport
| Cost index for Leeds–London carriers (1693–1702 = 100) | Cost index for London–Exeter coaches (1658 = 100) | |
| 1700–09 | 102.0 | |
| 1710–19 | 92.3 | |
| 1720–29 | 95.4 | 119a |
| 1730–39 | 96.8 | |
| 1740–49 | 82.4 | |
| 1750–59 | 62.3 | 73b |
| 1760–69 | 51.2 | 47c |
| 1770–79 | 50.9 | 46.3d |
| 1780–89 | 51.2 | 41.6e |
| 1790–99 | 46.4 | |
| 1800–09 | 44 | |
| 1810–14 | 47.8 | 35.7f |
| 1820 | 43.2 | 23 |
| 1825 | 36.2 | |
| 1838 | 30.9 |
a 1728
b 1757
c average of 1760–67
d average of 1776–78
e average of 1786–88
f average of 1810–14
Source: Gerhold (1996a, pp. 494, 508).
More than the manufacturing sector, transportation demonstrates how technological and institutional factors interacted and that any attempt to disentangle the effects of one or the other runs into what economists call “non-separability.” Gerhold (1996a, p. 506) concludes that the savings in transport costs were not due to the turnpike trusts as such, and that without them, new techniques such as MacAdam’s could have been adopted by some other form of road authority or even by the parishes. Yet Bogart’s demonstration of the importance of trusts suggests the strong synergy between institutional change and technological progress. In that respect the transport sector was a microcosm of the enlightened economy.
The most ambitious and costly project of the years of the Industrial Revolution was the construction of canals. In many respects, internal waterways were unglamorous projects. The technology involved was, in the main part, old and unspectacular. Canals were mostly designed for bulky, slow-moving cargoes and served mostly local transport needs, the average haul estimated at 26 miles or less. Much like turnpikes, they required parliamentary approval. They were also expensive to build and maintain, with a great deal of engineering ingenuity invested in the construction of aqueducts, embankments, bridges, locks, and tunnels. The early canals were still set up by landowners, who accounted for over 40 percent of all investment in them, but by the years 1780–1815 their share had fallen to 22 percent and manufacturers accounted for 15 percent (Hawke and Higgins, 1981, p. 233).
A number of canals have become causes célèbres in the economic history of the Industrial Revolution, above all the famous Bridgewater Canal completed in 1759, which connected the coal mines of the Duke of Bridgewater in Worsley with Manchester. It was a fairly small, local affair despite its great publicity, though it halved the price of coal in Manchester, and sent a powerful signal regarding the profitability and feasibility of canals. Much more consequential was the Grand Trunk (Trent and Mersey) Canal, completed in 1777 at the initiative of Josiah Wedgwood. It connected the east and west coasts of England and was designed and built, like many of the eighteenth-century canals, by the great engineer James Brindley. The Birmingham Canal, authorized in 1768 and completed in 1772, was a success, and by 1800 Birmingham had become “the Kremlin from which canals radiated in all directions” (Jackman [1916], 1962, p. 370). The challenge to connect the network that had grown in the Midlands to London was taken up in 1793, through the Grand Junction Canal, completed in 1805, which reduced the distance by almost 60 miles. The Chester Canal could boast the two huge iron aqueducts built by the miraculous Thomas Telford at Chirk and Pontcysyllte in eastern Wales.
Canal barges were slow: they were pulled by horses along towpaths. But for the purposes of hauling bulky and heavy loads such as coal, bricks, limestone, salt, timber, clay, and ore, they were of substantial importance to the continuing development of the economy. Inland waterways’ capacity almost tripled from about 1,400 miles in 1760 to almost 3,900 miles in 1830. The benefits were, of course, not evenly spread: canals were local affairs and served primarily local needs, and cooperation between adjacent companies was often lacking, to the annoyance of their customers. Canal companies were “extremely parochial” (Turnbull, 1987, p. 541). A contemporary author noted that “no towns have derived greater advantage of canals perhaps than Manchester and Liverpool” (cited by Harris, 1956, p. 158). All the same, their rather sudden surge in the closing decades of the eighteenth century was another mark of the determination of the age of improvement to apply the best useful knowledge they had in engineering to economic progress. Although the completion of a national network took decades, the gains of greater economic integration in the end accrued to the entire country. Indeed, it may well be questioned to what extent Britain could have taken advantage of its generous endowments of coal without the canal network.
A good example of the costs and benefits of canals was the construction of the Leeds–Liverpool Canal, commenced in 1770, which extended 127 miles across the Pennines. The project was interrupted when the company ran out of money between 1777 and 1790, and was only completed in 1816, although portions of the canal came into use as early as 1774. Under the inspired leadership of James Brindley and later his partner John Longbotham (another Smeaton pupil), it turned out a feat of engineering without precedent and amazed contemporaries. Because of the difficulty of the terrain, the canal required no fewer than 91 locks and a tunnel of 1,630 yards at Foulridge, yet it was highly successful and reduced transport costs between the two prime industrializing counties of England, Lancashire and Yorkshire, by as much as 80 percent (Baines, 1875). Like most canals, its costs were such that incorporation was required, but because the shares were large, only wealthy investors—mostly local landowners from Yorkshire – participated in the project and controlled its technical parameters.
The canal era was driven by the financial resources and entrepreneurial energies of local notables. They hired expert engineers such as Brindley, William Jessop, Thomas Telford, and James Green to design their canals, and overcame the often stubborn resistance of vested interests that stood to lose from the competition. The enormous technological advances that the engineering profession had experienced were epitomized in the magnificent Pontcysyllte aqueduct, which to this day is the longest and highest in Britain, built between 1795 and 1805 by Telford and Jessop, and the first such project to rely heavily on cast iron as a construction material.
British inland waterways went from triumph to triumph and many of them returned a healthy profit—until the trains came. With the arrival of the railroad, canal construction understandably slowed down and then stopped altogether. But economic historians have long seen canals as the “next best” alternative to railroads. They were slower, of course, froze over in cold winters, and lacked the technological excitement of the early trains. But speed mattered less for the cargoes that used the canals, and the notion that without the railroad industrial growth and economic progress would have ground to an early halt is no longer tenable. Canals, like turnpike roads, eventually turned out to be a costly dead end, though they long remained useful for some specific purposes. But the history of economic progress is inevitably studded with such “failures.” In a world of technologically driven growth, the road was littered with could-have-beens, the unavoidable cost of progress.
Coastal shipping has been termed the “Cinderella of the transport world” (Armstrong, 1996). It, too, was distinctly unglamorous, and devoid of melodrama and heroes. Yet that does not mean it was unimportant to the British economy. John Armstrong (1987, p. 176) has calculated that as late as 1910, coastal ships carried 59 percent of all ton-miles of internal freight, with the railroad picking up 39 percent and canals only 2 percent. For the first half of the nineteenth century the size of coastal shipping cannot have been less, although good numbers are lacking. Needless to say, ton-mileage is not the best measure of transport intensity, but it underlines the importance of coastal shipping as a cheap mode of hauling heavy and bulky cargoes over relatively long distances. In terms of hauling bulky and heavy goods such as cereals, bricks, sand, iron ore, and above all coal, it provided by far the cheapest mode and helped make Britain an integrated economy, contributing to regional specialization and efficiency. Baldwin’s London Directory published in 1768 listed no fewer than 580 places in England and Wales accessible by water—the majority of them being served by ports. Coastal shipping made energy-intensive industries, such as brewing, glass, bricks, and salt, possible in sites far removed from the mines, and it kept Londoners warm and London increasingly foggy by shipping in coal to be burned in private hearths. It was also one of the main sources of demand for the British shipping industry: in the first half of the eighteenth century the average annual shipment of coal from the northeastern ports such as Newcastle to London was around half a million tons a year, exceeding the total tonnage of Britain’s two bulkiest exports, grain and coal. To be sure, coastal voyages were shorter, yet these ships had to be seaworthy in every respect.
Coastal and cross-Channel shipping was technologically no less progressive than deep-water vessels, and in some advances in shipping technology coastal ships were the pioneers. As soon as steam power became feasible, it was adopted by coastal shippers, to reduce their dependence on the whims of winds and tides: as early as 1821, there were 188 steamers occupied in the British coastal trade, and their tonnage increased by a factor of ten in the next century. While their navigational demands were perhaps different than transoceanic shipping, they were far from trivial. Until the railroads came, it is hard to see how on many routes the heavy loads of fuel and building materials that industrialization and urbanization demanded would have been hauled around in the absence of coastal shipping. Canals were better suited for short-distance hauls, and in any case coasters were more of a complement than a competitor to canals. Coastal shipping also carried considerable passenger traffic, especially in the northern parts of Britain. The trips were slow, but more comfortable and cheaper than road transport.
One of the typical transport improvements of the age was the renovation of London Harbor and the construction of a dock system in a feverish construction scheme during the first years of the nineteenth century. The moving force behind the improvements in the West India Dock were London merchants with strong Caribbean connections such as George Hibbert and Robert Milligan. Parliament itself passed the Bill to authorize the construction (in 1799) and appointed the civil engineer William Jessop, John Smeaton’s star apprentice and protégé, and the chief engineer of the Grand Junction Canal, as the engineer in charge. John Rennie served as consultant to the project, and it employed one of the first high-pressure engines ever used in an engineering project for the dredging work, designed by Richard Trevithick himself (Burton, 2000, pp. 113–20). The smaller East India Docks were authorized in 1803, London Docks at Wapping were built between 1802 and 1805 (by Rennie as well), and the St Katherine Docks in the early 1820s, designed and supervised by Telford.
Railroads were the invention that par excellence defined modernity, both to contemporaries and to economic historians. Before the railroad, people had never been able to move at a speed exceeding that of a fast horse – and only a select few had experienced that. Travel by stagecoach was slow, expensive, and uncomfortable. Trains were faster, cheaper, and could reach places that previously were quite inaccessible. Like many of the other developments during the Industrial Revolution they were essentially democratic in that they made accessible services that previously had been confined to the rich and privileged. The majority of people who were to make use of passenger trains had had to walk to their destinations before 1830. The railroads, much like our own communication technology, shrunk the world—perhaps more so than any invention since the sailing ship. They were spectacularly visible and audible, and could be experienced at a low cost. The development of the railroad drove home, even to the most remote and peaceful rural regions of Britain, the message that the world was changing at an ever accelerating pace, and that in the long run no place would remain unaffected. But the railroads are equally interesting for the technological problems that they created and solved and the economic and institutional implications of the construction and operation of a project of unprecedented size and complexity.
Wooden tracks that minimized the friction created by pulling heavy cargoes on wheeled vehicles can be traced back to the early Middle Ages, and were quite widely used by British mines in the late eighteenth century. The idea of a smooth rail as a friction-minimizing surface for wheels had been used in coal mines since the eighteenth century, and Richard Reynolds (Abraham Darby II’s partner) substituted iron rails for wooden ones as early as 1768, but the use of iron rails on a wider scale would not have been possible without the puddling and rolling process. By the first decade of the nineteenth century, decades before the first successful locomotives, Britain was estimated to have 300 miles of railway track (Bagwell, 1974, p. 90). The first “general-purpose” railroad (built by Jessop) was the Surrey horse-drawn iron railway completed in 1805, and the famed “Mumbles” railroad established in Swansea was the first one to haul passengers. While no financial success, they indicated what this form of transport could do. The first use of steam power was on the Stockton and Darlington railroad (mixed horse- and steam power) in 1825. The conventional start of the railway age, however, is taken as the opening of the Liverpool–Manchester route in 1830.
The railroad was the unmistakable child of the Industrial Revolution. Its two central technological components were the steam engine and the iron rail. Neither of these was entirely new in 1830, when the first steam railroad was officially opened. The steam engine that propelled the locomotives was a high-pressure engine, developed in the first decade of the nineteenth century by engineers such as Richard Trevithick and Arthur Woolf. There were different kinds of machine that used “strong steam” in the parlance of the day, and Trevithick’s insight was to get rid of Watt’s separate condenser, which made for a lighter and smaller device. This engine was adapted by a brilliant father and son team of engineers, George and Robert Stephenson, to the specific purpose of creating a steam locomotive, using a revolutionary multi-tubular boiler. Other insights by engineers associated with new technologies also found their way into the complex technological issues that railroads involved, such as brakes, gears, axles, gauges, couplings, and springs. But the railroad also posed entirely new problems, none larger than the need to communicate rapidly over large distances to coordinate the movement of trains. The telegraph, the first large-scale technique to rely on electrical phenomena and thus just as radical and momentous an innovation as the trains it announced, developed about a decade after the railroad. If ever there was a case of technological symbiosis, this was it.
The railroad was the climactic achievement of British engineering competence. It had not much science or even formal mathematics underlying it. It was mostly designed and built by people with little or no formal education, but who had mastered a profound if informal understanding of what did and did not work, through a combination of natural talent and access to the right masters. The first models were built by Richard Trevithick, whose education was mostly provided by his own father and uncle in the Cornish mines (Burton, 2000, p. 28). George Stephenson had even less formal education, and he, John Blenkinsop (often credited with building the very first locomotive), and William Hedley, the designer of an intermediate proto-model of the locomotive known as “Puffing Billy, ” were all trained as practical mining engineers. Another railroad pioneer, Timothy Hackworth, similarly, was apprenticed to his father (a blacksmith) and he, too, worked at a colliery. The technical problems in the railroad were often hard and perplexing, but they were still of the kind that could be overcome with the traditional empiricist engineering skills that had stood British manufacturing in such good stead during the Industrial Revolution. It was, however, not a promising strategy for future technological advances.
The technical challenges of the railroad were matched by economic and institutional ones. For one thing, the construction of the network was by far the most costly and ambitious overhead investment project since the Pyramids. The questions of who could and would finance and manage it were raised in every nation that contemplated constructing a railroad network. Much like any other investment that has social overhead characteristics, there was an ambiguity about the role of the state. By the 1830s, Britain’s commitment to liberalism made its government hesitant to follow the model of continental nations like Belgium and Prussia where the government participated actively in the financing and construction of the new project. British railroads were financed by securities sold to the general public. But clearly pure laissez-faire in this sector was unrealistic. Parliament still had to approve each company, and some limitations were imposed on the free operation of railroad companies in the famous Railways Act of 1844 (drawn up by the young William Gladstone, then President of the Board of Trade), which limited the prices that railroads could charge and imposed safety regulations that remained on the books till the twentieth century. Even when it seemed that necessary coordination could be dealt with through private means, government support was needed. A case in point is the Railway Clearing House, set up in 1842 by a few companies to compute the net balance of each company’s dealing with others when passengers and freight were transshipped through more than one company. It also set technical standards that benefited all, such as the adoption of Greenwich Mean Time for railroad schedules and a variety of technical standards. In 1850 the members themselves had to initiate the Railway Clearing Act to compel some recalcitrant companies to participate in this scheme.
The railroad network turned out to be a huge investment project. To provide some context, consider this: in the 1820s, transport investment—roads, harbors, and canals—absorbed about 15 percent of British Gross Capital Formation, whereas in the late 1840s, the peak of the railroad boom, this figure jumped to a lower bound of 40–45 percent, with some of the higher estimates reaching 54 percent. Regardless of which estimate is more accurate, it was an order of magnitude above the already considerable costs of maintaining the existing superstructure. An analysis of the owners of the securities shows that a substantial number of these stocks and bonds were owned by merchants and landowners. Not necessarily the class that had been most active in leading the Industrial Revolution, but perhaps the one which indirectly had benefited most from it. The ability of the British economy to finance this huge project was truly impressive. Rates of growth of the railroad network are of course not very meaningful, -because it started from nothing: in 1830 Britain had less than 200 kilometers of railroad track. In 1850 it had completed 9,800 km, almost a third of the 30,000 km it ended up with in 1900. To carry out this project required a considerable reallocation of resources. In the feverish peak years of railroad construction, between 1845 and 1849, close to a quarter of a million men were employed, perhaps 4 percent of the male labor force (Mitchell, 1964, p. 323). Many of these workers were Irish immigrants; others may have been young males whose prospects in domestic manufacturing were bleak in the 1840s.
Part of the reason why the railway was so expensive was that the two leading engineers and builders, Robert Stephenson and Isambard K. Brunel, were both profligate with investor money. Stephenson had estimated that the London– Liverpool line would cost somewhat over £21,000 per mile, whereas the actual cost was over £50,000. Brunel had underestimated the cost of his London–Bristol line by a factor of 150 percent. A third engineer, Joseph Locke, was far more economical and was willing to build along steeper gradients, which saved costs. But until his approach became common practice, a lot of stockholder money was wasted. Moreover, railroad companies built extravagant, lavish, architect-designed stations, which cost far more than necessary. It may well be that those stations were intended to attract worried and reluctant passengers, but they added to the overall costs of construction (Rolt, 1970, ch. 1).
Other new institutional problems that emerged with the railroads had to do with the need to coordinate technical standards so that different railroads could connect to one another. Of these, the most notorious was the railroad gauge. Different railroad companies used different standards, with the obvious results that their trains could not use each other’s tracks, and freights and passengers had to unload and re-embark at terminal points. Two gauge standards emerged, the gauge used by George Stephenson on his Manchester–Liverpool line opened in 1830 (4' 8½") and the wide gauge preferred by the Great Western Line owner, Isambard K. Brunel. Long debates, sometimes known as the “gauge wars” ensued on the relative merits of each of these gauges. It was believed that wider gauges provided more stability to trains, and no less an authority than Charles Babbage supported this view. But in the end it became clear that differences between the two standards were less important than the cost of incompatibility and Parliament had to impose a standard. In 1845, a special commission recommended adopting the narrow (Stephenson) gauge for all new railroads and the last of the wide gauges, the Great Western, converted to the standard in 1892. The actual Stephenson standard selected was more “in the nature of a random draw from a variety of practices” but in this case diversity was costly and inefficient, and coordination by a government agency was necessary (Puffert, 2009, p. 48). It was one of the first historical cases of network externalities, and it became clear, even to the most laissez-faire Victorian, that arbitrating such disputes could be more efficiently done by an impartial outsider than by the market (Siddall, 1969).
The exact quantitative significance of the railroads to the overall development of the British economy will remain a matter of dispute. The locational patterns of manufacturing had already been set in 1830, and the appearance of the railroad did little in the medium term to change that—indeed railroad construction was determined by existing locations of manufacturing centers. Economists have developed the concept of social savings, the total net gain to society from developing a new technique compared to the next best technique available. They point out that the marked alterations in landscape and to some extent lifestyle that the railroad wrought were the means by which such social savings were achieved, not an addition to them. Gary Hawke (1970) applied this technique to the computation of the net social savings of British railroads in 1865, and estimated that the total savings came to 6–10 percent of GDP. In 1850 the net impact of the railroad on the economy was still not earth-shaking at around 2.5 percent of GDP (Gourvish, 1980, p. 34), with the bulk of the economic impact coming in the 1850s and 1860s. These numbers, however, are only as good as the assumptions made to compute them. The estimates depend rather crucially on the value placed on passenger comfort. Freight alone accounted for a social saving of only about 4 percent of GDP in 1865, while passenger services accounted for 1.5–6 percent depending on the value that passengers placed on comfort. The higher figure has been criticized quite effectively (Gourvish, 1988, p. 81). More recently, Foreman-Peck (1991) has revisited the 1865 computations by accounting for their impact on such second-round effects as enhanced labor mobility and higher capital formation, and found them to be more or less robust to any adjustments, though for the later decades the social savings may have been underestimated. Of deeper significance in revising these figures is research by Leunig (2006), who accounts for the time saved by passengers who could get to their destinations faster and more reliably, which he reckons at about 2 percent of 1865 GDP. Even that estimate could be argued to be a lower bound of the total economic welfare gain or “social surplus” created. If the passengers traveling third class had to walk before the time of the railroad, clearly the physical effort saved by traveling by train should have improved their well-being even more.
Ocean shipping was of course an ancient activity, but it, too, was wholly transformed in the period under discussion here. In 1700, Britain was still playing second fiddle to the Dutch as master of the oceans. The lucrative carrying trade (equivalent to the export of shipping services, an “invisible export”) was gradually expropriated by the British thanks to an aggressive mercantilist policy as expressed in the Navigation Acts. In the early eighteenth century British ships gradually began to out-compete others in the North Sea and Baltic trade, as well as the Atlantic (Ormrod, 2003, pp. 60–66). Part of the British success on the high seas was due to the strength of the British navy protecting British interests, its effectiveness in using privateering to disrupt Dutch and French trade during wartime, and the growing efficiency with which the British maritime sector financed, insured, and organized its enterprise. The Royal Exchange Assurance and its sister company, the London Assurance, were founded in 1720 and from then on British insurers competed successfully with the Dutch. The expanding colonial Empire increasingly required the services of British vessels. It is telling, indeed, that Dutch capitalists invested heavily in British shipping, a testimony to their efficiency. Yet the success of British shipping also reflects in large measure the growing openness of the economy, the skills of its seamen, and the capability of its mercantile institutions to organize and finance sea voyages, notwithstanding the often destructive mercantilist meddling with and limitations on trade. The Navigation Acts made sure that British shipping interests benefited from the carrying of these goods, but it stands to reason that even without any favorable policy, the rising efficiency of the British maritime sector ensured that it would have expanded pari passu with the growth of long-distance commerce.
The technological revolutions reached the ocean shipping sector relatively late. The application of steam to ships was slow, perhaps slower than the early experimenters had envisaged. For decades after the first famous crossing of the Atlantic by the Savannah in 1819, the oceans were crossed by sailing ships with auxiliary steam engines. One difficulty was that paddle wheels, the obvious form of propelling a steam-driven ship, worked poorly on the open waters, and it took decades until screw propellers were perfected in the 1850s. Another technological hurdle was that high pressure in marine steam engines was unusually hazardous, as salt in the water used to run the boilers tended to corrode the cylinders and lead to explosions. The surface condenser, which separated the water that cooled the condenser and the water in it, was developed in the 1830s but turned out to be a difficult problem (despite the efforts of some of the best minds in Britain, including William Thomson, later Lord Kelvin), and the ultimate victory of the steam engine on the oceans was not complete until the 1860s. The earliest steamers thus relied heavily on sails and their engines were auxiliary in nature, a hybrid technology if there ever was one. Yet by the 1830s, paddle steamers (“steam packets”) had established a regular service with the Continent and Ireland.
Although the full application of steam power to ocean navigation, then, did not occur until the second half of the nineteenth century, contemporaries saw its potential long before and used it as an illustration of the power of new technology to enforce the powers of enlightenment. Sadi Carnot wrote in 1824 that “The safe and rapid navigation by steamships may be regarded as an entirely new art due to the steam-engine. Already this art has permitted the establishment of prompt and regular communications across the arms of the sea, and on the rivers of the old and new continents. Steam navigation brings nearer together the most distant nations. It tends to unite the nations of the earth as inhabitants of one country” (1824, p. 4).
One striking feature of the entire transport sector is that, much as we saw in steam and water power, the “old technology” showed remarkable ability to reinvent itself not only because it was threatened by a competitor, but because the economy-wide phenomena of improved access to knowledge and better understanding of the details of the techniques in use affected traditional techniques such as sailing as well. Formal “science” was not yet the decisive factor, but improved engineering and materials, and the smoother flow of useful knowledge were. The fact remains that between 1820 and 1860, sailing vessels were completely redesigned after very slow change in the previous two centuries. By reducing the size of sails and increasing their numbers, sailing ships could be made more flexible and faster, and could be operated by fewer sailors, an important source of cost-saving. Metal replaced wood in fittings, copper sheathing was employed to prevent marine growth on iron hulls, and riggings were redesigned, and construction and operating costs declined. The best-practice knowledge of the day was applied to clear and well-defined technological problems. A typical advance in shipbuilding was John Scott Russell’s “wave-form” theory of ship design proposed in the 1830s, which, while flawed and no longer accepted today, led him to build more streamlined and therefore more efficient ships than ever before. British shipbuilders successfully competed with technological American clipper ships, which by 1850 were the top of the line of sailing ships.
The decline in the prices of materials coupled with an improved knowledge of their properties led to a radical change in the materials from which ships were made. When the ironmaster John Wilkinson launched the first iron-made vessel into the Severn at Coalbrookdale in 1787, a large crowd was attending, expecting it to sink like a stone. The iron ships were at first built more like wooden ships, with transverse iron frames that did not take advantage of the inherent strength of iron plating. The entire art of building ships had to be relearned. By the middle of the nineteenth century iron had replaced wood as the main shipbuilding material. The use of iron (and later steel) in hull construction once and for all removed the constraints on the size of ships that wooden construction had imposed from the earliest days of shipbuilding. It has been estimated that ocean freight shipping costs fell by 0.88 percent per year between 1811/30 and 1852/58—and most of this took place through small technological improvements in the design of wooden sailing ships or through the improvement of harbor facilities or the greater reliance on tugboats (Harley, 1988, p. 861). The eventual disappearance of the great sailing ships from the oceans should not obscure the efforts and successes in improving old and ultimately “unsuccessful” designs. Techniques should not be overlooked even if the future was to determine that many of these old techniques were moribund. The wooden sailing ship, the design of which had changed but little between 1650 and 1850, all of a sudden underwent dramatic productivity change, as new knowledge was brought to bear on an old technology.
The consequences of the advances in ocean shipping and the concomitant decline in shipping costs were far-reaching. The growth in the volume of international trade before 1850—much faster than the growth in output—must be largely chalked up to the decline in transportation costs. But better ships had unexpected results not wholly reflected by shipping costs alone. At the beginning of the nineteenth century, it could take as long as two years for a letter from Britain to Calcutta to receive an answer, in part because the Hooghly river was hard to navigate upstream due to monsoon winds. By 1840, the one-way journey around Africa had been cut to six weeks, as auxiliary steam had solved the problem. At the same time a single steam-propelled gunboat, the Nemesis, and her unexpected ability to sail up the Yangtze river and blow away Chinese ships with her superior guns, determined the outcome of the Opium Wars and provided a painful illustration of the gap that by that time had opened between Western and Eastern technology.