The Tycoons: How Andrew Carnegie, John D. Rockefeller, Jay Gould, and J. P. Morgan Invented the American Supercompany

by Charles R. Morris

  To Thomas Aquinas, the universe was still a person; to Spinoza, a substance; to Kant . . . a categorical imperative; to Poincaré, a convenience; to Pearson, a medium of exchange. The historian never stopped repeating to himself that he knew nothing about it. . . . He saw his education complete, and was sorry he ever began it. As a matter of taste, he greatly preferred his eighteenth-century education when God was a father and nature a mother, and all was for the best in a scientific universe.

  America’s raucous economic successes were not much consolation. As a youthful Walter Lippmann put it, modern industry “is the great fact in our lives, blackening our cities, fed with the lives of our children, a tyrant over men and women, turning out enormous stocks of produce, good, bad, and horrible.” And although all Americans on each Fourth of July duly celebrated “the huddled masses yearning to breathe free,” that same “wretched refuse” from distant teeming shores was clearly making a terrible mess of big eastern cities. The crime rates and disease in Boston’s and New York City’s slums were horrific.

  But a deus ex machina, quite literally, was on hand for the rescue. When U. S. Steel was founded, Charles Schwab stressed that the biggest companies were run by specialist managers trained in “the science of business”: “Nothing is left to chance. Every step of the process is carefully worked out in advance. All waste is cut off.” Lippmann, who was a reliable weathervane of the period’s intellectual fads, enthusiastically echoed Schwab: “American business has been passing through a reorganization so radical that we are just beginning to grasp its meaning. . . . The scope of human endeavor is enormously larger, and with it has come . . . a general change in social scale.” But the “new business world has produced a new kind of business man. For it requires a different order of ability to conduct the Steel Trust, than it did to manage a primitive blast-furnace.” Trust-busters, Lippmann said, failed to understand that the right size for a business was a matter for “experts in the new science of administration. . . . The fact is that administration is becoming an applied science, capable of devising executive methods capable of dealing with tremendous units.” In Edward Bellamy’s utopian 1887 novel, Looking Backward, social strife disappeared once all production was put in the hands of a “single syndicate. . . . The Great Trust.”

  From a chair in academia, or from a journalistic desk, giant corporations like Standard Oil or U. S. Steel took on the polished, quietly humming appearance of the awesome Corliss engine that had towered like a brooding god over the Philadelphia Exposition a generation before. The scientific approach to business was underscored by the star attraction of the 1904 St. Louis Exposition, the Pennsylvania Railroad’s display of a working locomotive testing plant. Mammoth overhead cranes could swing even the largest locomotive onto mechanical rollers, where it would power up and roar away at top speeds—schedules would announce when visitors could watch a locomotive running at, say, seventy miles per hour—while teams of technicians carefully measured temperatures, fuel consumption, tractive resistance, and pulling power, stopping the run from time to time to change a part or adjust a setting.

  The notion of businessman-as-scientist flowed directly from Pearson’s insistence, in his Grammar of Science, that science was primarily about method. The scientist strove for pure objectivity, impersonal, value-free. He proceeded by “the careful and often laborious classification of facts, in the comparison of their relationships and sequences, and, finally, in the discovery . . . of a brief statement, or a formula, which . . . is termed a scientific law.” Science-as-method, Pearson wrote, “claims that the whole range of phenomena, mental as well as physical—the entire universe—is its field. . . . every phase of social life, every stage of past or present development is material for science.” There was, in short, no need to jettison the promise of Progress and American exceptionalism, but the path to the shining City, instead of being lighted by God, would be revealed through Science.

  Scientific Management in action: The Pennsylvania’s locomotive testing laboratory at Altoona, Pennsylvania. Previously, it had been an exhibit at the 1904 St. Louis Exposition.

  This was heady stuff. The success of marginalist economics seemed to buttress Pearson’s claim. Just as a few simple laws choreographed the freely colliding molecules of a gas, correct prices arose from the activity of countless atomized market participants obeying simple canons of self-interest. The fledgling study of sociology jumped on the statistical dogcart; when the American Sociological Society was founded in 1905, it was open only to “scientific” practitioners. Sociology was expressly framed as a science of “social control,” teasing out the laws of individual interactions that created a “social equilibrating apparatus.”* Even Henry Adams tried his hand at a “Dynamic Theory of History,” hopeful of discovering the tidal laws governing the rise and fall of nations. John Dewey was confident that schools could be run like “great factories” to churn out the self-reliant citizens to people his vision of a liberal democracy.

  The cult of the expert was born. Dewey said that science aimed at “the transformation of natural powers into expert, tested powers.” The Pennsylvania Railroad’s publicity material at St. Louis emphasized that the test routines were rigidly specified and minutely directed by a Purdue professor, one F. M. Goss. There was a core of truth here: many big companies were indeed building research laboratories and driving toward scientific quality control and product development; but the relationship between businessmen and scientists—even at the Pennsylvania—was arm’s length and prickly, as it largely remains today. For journalists and intellectuals, however, who often knew little about business and less about science, the scientific expert transmuted into a kind of wizard. The historian Theodore Porter has noted that the Pearsonian version of science was “ideally suited to American democracy. Social scientists . . . could disarm suspicion that their advice was self-interested by intoning the phrase scientific method.”

  So when Frederick W. Taylor proclaimed that he had discovered the principles of “Scientific Management,” his audiences went into a collective swoon.

  What Did Taylor Do?

  Frederick Winslow Taylor, born to a wealthy Philadelphia family in 1856, was a prodigiously gifted young man, vigorous and outgoing, a natural leader, and a good student with a strong bent for mathematics and physics. He was also a superb athlete, for he and a friend won the U.S. Open tennis doubles championship in 1881. After graduating from Phillips Exeter Academy, he passed up college to become an apprentice machinist in a local company owned by a friend of the family. After four years learning his trade—a period which he later claimed gave him special insight into the minds of ordinary workers—Taylor became a subforeman at Philadelphia’s Midvale Steel in 1878. He proved to be a hard “driver” style of manager, imposing monetary fines for ruined work or waste, and experimenting with various piece rate pay systems. He spent more than ten years at Midvale, rising through series of promotions to chief engineer, while he earned his mechanical engineering degree at night. Most of the basic themes of his subsequent work can be traced directly to his Midvale experiences, including his lifelong hostility to “soldiering”—the tactic of manual workers to settle into comfortable group-enforced output norms. He also proved to be a brilliant mechanic, and was awarded a number of patents for improved machine tool designs.

  The years Taylor was at Midvale, roughly the decade of the 1880s, marked a pronounced business scale shift from local toward regional or national organization modes. Railroads led the way, in the process surmounting management challenges of an entirely new scale—thousands of miles of roads, millions of shipments, far-flung construction and maintenance activities, tens of thousands of employees. The railroads’ drive to standardization, cost management, and quality control forced comparable adaptations at their suppliers, like Carnegie Steel, Westinghouse Airbrake, Baldwin Locomotive, and Pullman Sleeping Car. Holley evangelized the steel industry on the heavy costs of sloppy control—the unscheduled shutdown of a big blast furnace could
cost a small fortune—while Carnegie methodically mechanized away most of the old steel craft trades. Comparable developments took place in other railroad-enabled industries, like flour, sugar, and chemicals, and in the new mass distribution companies, like Montgomery Ward, the food store chains, and the big department stores. As operating scales outran the personal reach of top managers, there was a proliferation of paper-based control systems—departmental cost-tracking, standard paying and receiving systems, data tabulation and performance reporting. Office furniture, filing systems, forms, and typewriters became important industries, and office towers recarved urban skylines.

  But outside of the biggest or most advanced companies, the penetration of systematic management was spotty at best, and more often almost nonexistent. Especially in midtechnology mechanical industries, operations were typically a mess. The manufacturer and reformer Henry Towne, in 1886, told the American Society of Mechanical Engineers (ASME) that “the management of works is unorganized, is almost without literature, has no organization or medium for the interchange of experience. . . . The remedy . . . should originate from engineers.” The typical manufacturer often had minimal local competition, was rarely held to exacting quality standards, and had grown mostly by hiring more and more craftsmen in the old artisanal tradition. Internal contracting, or the use of independent contractors within the factory on a fixed piecework basis, was still common. And even in bigger, well-managed companies, midsized craft operations, like the machine shop in a steel plant, still often ran as if they were independent companies. Shop operations were Taylor’s sweet spot, especially the job shop, where there was likely to be considerable variation from one job to the next.

  There was an obsessive streak in Taylor, much like John Hall’s. A half century before, Hall had spent years on the challenge of machining interchangeable precision parts, driving down to each microlevel obstacle, and attacking and mastering them one by one. Taylor took on shop management in much the same way. Getting control implied standardizing every aspect of production—the quality of the machines and the cutting tools, the speed of the tool, the depth of the cut, the rate of the feed, the sequence of operations—details that Midvale, like most shops, left up to the foremen or the individual machinists. Step by step, Taylor isolated the critical performance variables and drove to a best-practice standard. As his system took shape, cutting tools were maintained in a central tool room, a specialist team oversaw the belting,* a planning group laid out production schedules, jobs were allocated with an instruction card that specified machining sequences and tolerances, materials were charged out to each job, and job and time cards tracked each machinist’s performance. The last step in the process was a piece rate, with the punitive feature that the piece rate dropped at lower levels of production. To set the rates, Taylor introduced stopwatch timing of the detailed operations. If Taylor’s subsequent practice is any guide, he confined his time study to the best men, and his timing was approximate at best. In one Midvale example that survives, the men would have had to nearly double their production to earn their former wage.

  Taylor’s obsessiveness was matched by a streak of grandiosity. Once he had the machine shop in order, he pushed to apply the same techniques to the entire company. A sparse record suggests that outside of the machine shop he worked only with laboring units, and, since there was no machinery involved, concentrated only on time studies and piece rates. At some point, he became enamored of the idea that all human actions could be engineered like a machine. As he put it some years later, “every single act of every workman can be reduced to a science.” But it required trained experts:

  [A] man who is fit to handle pig iron as a regular occupation is . . . so stupid and so phlegmatic that he more nearly resembles in his mental makeup the ox than any other type. . . . He is so stupid that he . . . must consequently be trained by a man more intelligent than himself into the habit of working in accordance with the laws of this science before he can be successful.

  Even the most intelligent workers needed the help of experts, however:

  [I]n the higher classes of work, the scientific laws which are developed are so intricate that the high-priced mechanic needs (even more than the cheap laborer) the cooperation of men better educated than himself in finding the laws . . . and training him to work in accordance with them. . . . [I]n practically all the mechanic arts the science which underlies each workman’s act is so great and amounts to so much that the workman who is best suited to actually doing the work is incapable, either through a lack of education or through insufficient mental capacity, of understanding this science.

  Which reveals another of Taylor’s traits, a penchant for pompous mystification. One imagines that old “Big Bill” Rockefeller, John D.’s medicine man dad, would have saluted a kindred spirit.

  After Taylor left Midvale in 1889, he bounced from job to job for another decade—a paper company, a ball bearing company, the Cramp shipyard (that built Pierpont Morgan’s Corsair), an electric motor business, then back to the ball bearing company. He consistently proved himself an outstanding plant manager, mostly through hard-driving piece rate systems and ruthless winnowing of workers who didn’t perform at the top level. (In his own favorite example of pig iron loaders, he set the piece standard so high that only one of every eight men could meet it. So he got rid of the rest and replaced them with far fewer “first-class men.”) Along the way he continued to polish up his ideas for shop management, which, although always intelligent, tended toward the fussy and overcomplicated. In his ideal machine shop, for instance, a machinist would report to eight different functional foremen. His presentations at the ASME, especially on piece rates, began to attract a small band of acolytes, including Henry Gantt, creator of the famous “Gantt chart,”* who joined his team at Midvale.

  One of his new assistants was Sanford Thompson, whom Taylor had met at the paper company and whom he hired to pull together his Midvale time studies for publication. (Taylor was wealthy enough to pay for staff out of his pocket; besides his family resources, he was earning a growing royalty stream from his tool inventions.) When Thompson discovered how rudimentary and inconsistent Taylor’s time studies had actually been, he and Taylor agreed that he should start over from scratch. Over the next six years Thompson performed detailed time analyses of construction site workers in eight trades running from “excavation” through “rock quarrying.”

  Thompson’s results, together with a manual of timing techniques (including how to conceal the stopwatch from workers), form a major section of Taylor’s 1903 text, Shop Management. It is a splendid example of sham science and spurious specificity run riot. In the “barrow work” sub-trade, the dirt-moving planner could assume that it would take a man 1.948 minutes to load a barrow with loosened clay, at a rate of 0.144 minutes per shovelful, while sand required only 1.240 minutes at a rate of 0.094 minutes per shovelful. Starting the barrow took 0.182 minutes, wheeling it 50 feet on level ground 0.225 minutes, and dumping and turning 0.172 minutes. And so on. Those results are reduced to a “general formula for barrow work” where “a = time filling a barrow,” “b = time preparing to wheel,” etc., to arrive at:

  B = (p+[a+b+c+d+f+(distance hauled/100)(c+e)]27/L)(1+P)

  There are additional formulae for filling the shovel and throwing the material, or filling the shovel, walking, and then throwing the material, and helpful tables for calculating throwing time based on the distance and the height thrown. Where the vertical throwing distance is four feet, and the horizontal five feet, it takes 0.073 minutes to fill the shovel and 0.031 minutes to throw the material. Throwing time rises to 0.043 minutes (or by seven-tenths of a second) if the vertical is increased to six feet while holding the horizontal constant. Taylor points out that the time for filling a shovel is independent of the distance thrown, but does vary with the kind of material, so the tables provide different values for various earth types. There is also a handy table of equations for deriving times of operations that are too quick t
o capture with a stopwatch, but for the equations to work “the number of successive elements observed together must be prime to the total number of elements in the cycle.” Clearly, any shoveler who aspired to become a supervisor had a lot of book work ahead of him.

  The silliness of it all is betrayed by the capital P, the last term in Taylor’s long formula above. The P represented the time a worker needed to rest, or was consumed by something other than full-bore production. It was always a large, round number. In one extensive assignment covering many different jobs, the P ranged from 25 percent to 75 percent, obviously overwhelming the three-decimal time-study tables. Where did the P come from? In fact, it was a best guess, but when pressed, Taylor fiercely stuck to his guns: P was never arbitrary, but was based on “scientific investigation, a careful, thorough, scientific investigation of the facts.” When a congressman suggested that traditional piece rates were also based on a foreman’s long observations, Taylor insisted that “The one is guesswork, while the other is a careful scientific experiment.”

  Taylor’s first and only full-time consulting assignment came in 1898 at Bethlehem Steel, which was experiencing serious production problems in its armor business. One of the Bethlehem senior executives had worked with Taylor at Midvale and admired his piece rate ideas, and so arranged for a presentation to Bethlehem management. Taylor stressed that it could take up to two years to install a full-blown piece rate system, because all the other elements had to be in place before he could conduct useful time studies. The board was enthusiastic, and Taylor began work in the spring. On its own terms, the engagement was a failure, and Taylor was fired two years later. Ironically, it was also the occasion of his greatest contribution to machining technology, the discovery of high-speed tool steel.