For the most part, electrification was a blessing to workers, making factories brighter, more pleasant, and safer, although electricity also brought the previously unknown dangers of shock and electrocution into the factory. Immigrants who came in contact with the force of electricity for the first time were the prime victims: “A newly landed Croatian lad of seventeen was killed by fooling with a switch with wet gloves on, watching the sparks fly.”16 Overall, however, electrification and safety went hand in hand. Belts, gears, and shafts were the main sources of factory accidents, posing a constant danger to workers’ fingers, arms, and lives. The switch to unit drive eliminated the jungle of belts and shafts and the accidents associated with them. Electrical machines also stirred up less dust, making the air cleaner and working conditions healthier. The displacement of gas by electric light further reduced humidity and improved oxygen in factories, while making acid fumes a matter of the past. And, increasingly automatic machines eventually lightened toil. It is thus no wonder that factory electrification for the most part was welcomed by workers (chapter 6). Indeed, in the period 1926–56, when the first comprehensive statistics on injury frequency rates were collected, the average number of disabling injuries in manufacturing was halved, as was also the case in mining.17 As one factory worker at Henry Ford’s River Rouge plant marveled in 1955, “Automation has saved me.… If I had to lug those heavy blocks into position like I used to I could not last till I was 65. Now I expect to be working till I am 80.”18 His only complaint was having put on thirty-three pounds since being aided by machinery.
As work gradually became less physical demanding, industrial hygienists of the 1950s and 1960s speculated that many injuries from the lifting, handling, and unloading of goods would become historical concerns. In textile weaving, safety devices were introduced on looms, automatically disconnecting the machine in case of any accident. And one Ford plant is reported to have experienced an 85 percent reduction in cases of hernia after the installment of automatic machines.19 Automation, according to the BLS, meant that man no longer had “to pace himself to the rhythm of the machine, a rhythm which may be an unnatural one and result in tension and possible accidents.”20
Machines meant the end of the most hazardous, dirty, and backbreaking jobs. For millennia, agriculture had been most people’s preoccupation. Now, in less than a century, technology had shifted the bulk of the American workforce from farms to factories and offices. Table 1 illustrates the remarkable occupational evolution of the American workforce: between 1870 and 2015, the share of working people in agriculture declined from 45.9 percent to 1.0 percent. While that share was falling even before 1900, this decline was due to the relatively rapid expansion of blue- and white-collar jobs. In terms of total employment, the agricultural sector saw its peak in 1910. Thereafter, it lost jobs in every decade. The main reason why workers left the farm, as we shall see, was more lucrative job opportunities in the cities. This, in turn, sparked incentives to mechanize. Tractor use slowly expanded after World War I, but the main burst of growth began in the late 1930s, and by 1960 80.0 percent of American farms had tractors, up from 16.8 percent in 1930.
The tractor alone can account for much of the reduction in labor requirements in farming. According to estimates by the Department for Agriculture, in 1960 the tractor had saved 3.4 billion man-hours in field operations and caring for draft animals, representing the equivalent of 1.74 million farmworkers.21 By that time, the agricultural sector had experienced a net reduction of 5.7 million farming jobs since its peak. Though labor-saving estimates for automobiles and trucks are unavailable for 1960, in 1944 they saved more than 1.5 billion man-hours in hauling and travel time and another 1.1 billion man-hours that would have been devoted to taking care of horses and mules.22 At the same time, rural electrification relieved workers of tedious tasks such as the milking of cows by hand, and irrigation became less laborious with the arrival of electrical pumps. Like motorized vehicles, electrification virtually eliminated the need for workers in some domains: “By the middle 1920s the local Campbell Ice Cream and Milk Company had electrical processing equipment connected by pipes, so that the milk never had to be handled directly or exposed to the air, but could be pumped from one stage to the next as the machines separated the milk from the cream, heated the raw milk to pasteurize it, homogenized it, cooled it to near freezing, and then bottled it.”23
Meanwhile, in mining, mechanization made inroads into the heavy manual work of loading coal onto electric cars in the 1920s, described at the time as “the most widespread form of drudgery existing in industry today.”24 In a mere decade, the volume of mechanized loading in coal mines expanded by a factor of twenty. And mechanized loaders were entering metal mines as well. In the Michigan copper mines in the early 1930s, it was reported that the adoption of “mechanical loaders and scrapers has replaced shovelling in large measure.”25 And as hard physical labor in the mines and the factories was gradually transferred to electrified machines and backbreaking toil on the land was taken over by motor power, workers shifted into the comforts of air-conditioned offices: as shown in table 1, people leaving agriculture after 1940 mainly took up work in offices.26 This was largely due to the spread of office machines, not despite them. The time saved by the use of the great number of typewriters and adding and calculating machines was surely significant, but without them many tasks would have been too laborious to be economically feasible on a large scale. Much of the work done by office machines would probably not even have been done had the machines not been invented. As Harry Jerome wrote, “If letters were all written by hand, and computations all made by laborious and expensive human effort, there would be a marked shrinkage in the volume of correspondence and computation considered necessary and economical.”27
In other words, machines were responsible for relieving workers of the most dangerous and physically demanding tasks as well as for creating new and more pleasant ones in electrified factories and air-conditioned offices. Economist Robert Gordon has calculated that the share of the workforce engaged in jobs that can be deemed physically challenging and dangerous fell from 63.1 in 1870 to 9.0 percent in 1970.28 Of course, such estimates inevitably understate the demise of unpleasant work because the content of many unpleasant jobs changed for the better as well. As Gordon notes, “One only need contrast the 1870 farmer pushing a plow behind a horse or mule, exposed to heat, rain, and insects, with the 2009 farmer riding in the air conditioned cab of his giant John Deere tractor that finds its way across the field by GPS and uses a computer to optimally drop and space the seeds as the farmer reads farm reports and learns about crop prices on a fixed screen or portable tablet.”29
More Jobs, Better Pay
Technology not only made jobs less hazardous and physically demanding, it also led to better-paying jobs. In the period 1870–1980, hourly compensation kept track with labor productivity (figure 9). Of course, factors other than technology also affected compensation. Though this book is about long-run trends in living standards rather than short-run fluctuations, a few arguably significant variables warrant some discussion. Part of the reason why real wage growth was rapid in the early twentieth century has been attributed to the rise of welfare capitalism in the 1910s and 1920s, which gave a boost to workers’ wages as companies sought to retain their employees. With more and more capital tied up in machines, the skills required to operate them naturally became increasingly valuable. At the Ford Motor Company, the cost of constantly training new workers, as many others left the company for better jobs elsewhere, prompted action. To keep its workers, the company introduced the five-dollar-a-day wage, effectively doubling the wages of the employees working in its factories. Since Ford accounted for almost half of the American production of automobiles, increasing the wage to five dollars a day can justly be regarded as “the most dramatic event in the history of wages.”30 In addition to raising wages, Ford also introduced a new welfare program for its workers, and companies like Procter and Gamble, Genera
l Electric, and Goodyear Tire soon followed suit, institutionalizing similar programs that returned some of the productivity gains to employees in terms of better wages, medical services, pension plans, and so forth. When the BLS surveyed 431 companies in 1917, it found that almost all of them had some form of welfare-capitalist scheme in place.31
FIGURE 9: Real Gross Domestic Product per Hour and Real Hourly Compensation of Production Workers, 1870–2015
Sources: See appendix, this volume.
However, what has been described as corporate altruism is more accurately referred to as corporate paternalism. Welfare programs were rarely unconditional. The Ford Motor Company, for example, set up a Sociological Department to advise its employees on improving their lifestyles, and department staff members visited employees’ homes to inspect them for cleanliness and make sure that workers were in fact married, as the company required. It was not uncommon for Ford employees to get young amateur actresses to play the role of a loving wife during company visits.32 But welfare capitalism nonetheless raised workers’ expectations of what employers must provide. As Louis Hertz’s The Liberal Tradition in America illustrates, the commitment to Jeffersonian individualism, small government, and strong property rights is deeply ingrained in American culture.33
Quite possibly, welfare capitalism helped pave the way for what the historian Jefferson Cowie has called the “great exception” in American political history—that is, the era of “collective economic rights” created by the New Deal. New Deal legislation evidently did much to shift the balance of power between capital and labor.34 The National Labor Relations (“Wagner”) Act of 1935 guaranteed the right of employees to organize into trade unions and bargain collectively for better conditions. It also provided mechanisms for solving disputes, such as the establishment of the National Labor Relations Board to mediate between employees and employers. Other legislation targeted compensation more directly. The National Industrial Recovery Acts of 1933 and 1935 authorized the president to prescribe a limited code of fair competition, including setting minimum rates of pay. And the Fair Labor Standards Act of 1938 introduced the forty-hour workweek for many Americans, while mandating employers to pay workers for any overtime hours.
Welfare capitalism and New Deal legislation surely affected compensation. But needless to say, these variables cannot account for the trajectories of real wages over a century. The strengthening of unions might, of course, also have affected wages over the long run.35 It is true that most studies suggest that unionized workers earned higher wages than their nonunionized counterparts as late as the 2000s, when much of their power had already faded. Labor unions clearly played a role in raising pay for their members and in giving them political voice. But in the end, the bargaining power of unions depends on the value of the skills and knowledge of the workers they represent. The case of the telephone operator illustrates this point vividly. When operators went on strike in 1968, it had virtually no effect at all, as automation kept the system running. The strike was barely noticed, apart from the headline “Automation Keeps Struck Phone System Working.” Citizens trying to call friends and relatives in different parts of the country were for the most part able to connect themselves. As one contemporary observer pointed out, “If a telephone strike had taken place a dozen years ago it would have been difficult and often impossible to place a long-distance call in the United States.”36 In contrast, automation allowed a few executives and supervisors to fill in for the 160,440 strikers. The simplicity of the automated system meant that the other employees had few technical difficulties taking over from the operators. In cases where the skills of a union’s members became redundant, the union lacked the bargaining power to make much of a difference. The best it could do was negotiate exit deals for its members. When longshoremen saw their skills made obsolete by containerization, for example, the longshoremen’s union achieved nothing more than financial compensation and retraining for the workers who lost their jobs to machines. And quite naturally, as the occupation of longshoremen diminished, so did the clout of the union.
On balance, it seems safe to conclude that the strengthening of unions following the Wagner Act can account for part of the growth in real wages relative to those before 1930. But the wages of textile workers grew steadily even during the closing decades of the nineteenth century, when textile unions were weak.37 And as we all know, whether people unionize or not, wages can grow over the long run only if workers continue to become more productive. Figure 9 shows how hourly wages rose together with output per worker until the 1970s. In America, like in Britain, this pattern emerged in the absence of organized labor and without any significant government intervention. The rising tide of people’s wages gives additional weight to the view that the first three-quarters of the twentieth century was a time when much technological change was of the enabling sort. As Gordon notes:
Some part of the explanation of rapid real wage increases before 1940, particularly between 1920 and 1940, may be attributable to the end of mass immigration and the encouragement of labor unions by New Deal legislation. But ultimately it was technological change that drove real wages higher. Part of this was compositional—new machines that pulled, pushed, carried, and lifted shifted the composition of employment away from the common laborer to operatives doing specialized albeit repetitive tasks and to new layers of supervisors, engineers, and repairmen to plan the layout of the machines, train new workers, and tend the machines. Firms began to raise pay to reduce turnover, for the assembly line could be slowed if an experienced worker quit and was replaced by someone who could not initially keep pace. Much of this shift in the nature of employment was created by the rise of the automobile industry and its assembly line method of production and is symbolized by the contrast between the dark, satanic steel mills of the 1870s and the smoothly running Ford and General Motors assembly lines of the 1920s.38
As discussed, the Second Industrial Revolution spawned new tasks for labor. The general-purpose technologies of the century boosted productivity and employment, while reducing unemployment.39 Technology also increased workers’ earnings capacity by bringing steam to low-productivity sectors like farming and by allowing more citizens to shift into more productive and better-paying jobs. The growth of the electrical industry offers a remarkable illustration of the ingenuity of inventors, the entrepreneurial spirit of enterprise, and the fluidity of the American workforce. Like the automotive industry, which overtook the railroads as America’s largest industry in 1940, the electrical industry became a significant operation. Together with its supplier industries, it supported millions of Americans. As electrical appliances poured into homes, relieving the American housewife of many burdens, mass production created a host of previously unimaginable occupations and industries. An early survey of new industries in 1905 takes note of the high pay in all of the electrical industries, explaining why not a single strike of any serious magnitude had been recorded. Around that time, the electrical industry was still relatively minor, providing employment for some forty-six thousand people.40 However, in the succeeding decades the mass production of telephones, radios, washing machines, refrigerators, electric irons, and so forth required more and more operators to meet Americans’ growing appetite for new consumer goods.
Old industries gave way to new ones. Employment in the automotive industry, for example, peaked nearly a century after textile employment. But even old industries continued to expand, as mass production made many consumer goods available to a growing percentage of the population.41 Crucially, workers in these industries benefited from production technologies that allowed them to earn better wages. In several separate case studies, the BLS found that the introduction of new machines led to the creation of new tasks and better-paying jobs.42 The growth of these industries was also the best unemployment insurance people could get. A blue-collar worker who lost his or her job had many more options at a time when semiskilled jobs were abundant. Clearly, technology did cause some occupations to
vanish—like those of lamplighters, elevator operators, laundresses and so on—yet these jobs employed only a fraction of the workforce relative to the new machine-aided occupations that emerged.
Of course, as noted above, millions of farm jobs disappeared. So what happened to all the farm laborers? Following publication of an influential 1967 paper by the economist Richard Day, it was long believed that an explosion of labor-saving technologies in agricultural production forced workers to leave the countryside.43 Indeed, Otha D. Wearin, an agricultural columnist in Iowa, wrote in 1971: “The productive capacity of power machinery has greatly reduced the farm population. Occupied farming units have become fewer and fewer, and farther and farther apart, as producers with power machinery reach out for more and more land to justify their investment. Country churches, country schools, country society and small country towns have suffered. In fact, many of them have completely disappeared.”44
The decline of rural institutions and communities was partly offset by the automobile, which made people more mobile. But there can be no doubt that some rural communities suffered. However, mechanization of agriculture was rarely the reason. We now know that Day hugely overstated the role of mechanization in cotton harvesting, claiming that cotton in the Mississippi Delta was entirely harvested by machine in 1957, when in fact only 17 percent of it was. A recent study of the cotton harvester shows that 79 percent of the decline of hand picking was due to wages rising more rapidly elsewhere.45 Rather than being pushed out of the farms, laborers were pulled out by better-paying jobs in the cities. In fact, the mechanization of agriculture was in large part the result of cheap labor leaving the countryside. The economists Richard Hornbeck and Suresh Naidu have demonstrated that the exodus of labor from the rural areas prompted farmers to invest in mechanization.46
The Technology Trap Page 23