The year 1919 should have been a very good one for E. I. du Pont de Nemours and Company. World War I had blessed the 117-year-old, Delaware-based explosives manufacturer with the kind of booming growth that only an arms-maker in a global conflagration can dream of. The company had then leveraged this wartriggered glut of cash, plants, and personnel to accelerate its long-planned expansion into dyes, paints, plastics, and other new chemical-related ventures. Du Pont believed that transforming itself from singleindustry explosives firm to diversified chemical combine would prove a winning strategy in the post-war era.
Except it didn’t work out that way. Du Pont’s new lines performed poorly. Profits in plastics plunged. Despite the investment of millions of dollars, the dye operations lost money. So did the company’s chemical ventures. Paints and varnishes were a particularly troubling area. Du Pont had expected that the economies of scale made possible by its large size and vertical integration would quickly propel it to leadership in an industry made up mostly of small, nonintegrated firms. And in fact sales had soared but so had losses.
In 1919, gross paint and varnish sales rose 38%, to $4 million, from the previous year, but losses, nearly half a million dollars, were up an even greater 52%. “The more paint and varnish we sold, the more money we lost,” noted an internal report. Du Pont’s new ventures were lagging not only the company’s own targets, but the performance of its competitors in those industries, as well. 1919, for instance, was one of the most profitable years ever for many of the smaller paint companies that Du Pont had expected to surpass. The problem seemed to be selfinflicted.
But what was it? To find out, Du Pont formed a committee to assess the situation. It was made up of junior executives with operational responsibilities in the company’s major departments. The committee began by examining the new lines’ marketing shortcomings, which were considerable. The closer they looked, however, the more their focus broadened. Finally, after six months of study, these young middle managers arrived at a far reaching and unexpected conclusion: The root cause of the problem was the core structure of the company itself.
The solution they suggested was no less sweeping. Du Pont’s centralized, functionally departmentalized structure—the same structure which the company had adopted with great care less than two decades before, which had served Du Pont so well through World War I, and which was still considered the state of the art for a great vertically integrated industrial enterprise—should be blown up. In its place, the junior executives proposed that the company adopt a novel, decentralized configuration – one that was without precedent in an American industrial firm.
Consolidation and Centralization To understand fully the nature and significance of Du Pont’s historic transformation, it helps to start at the beginning. For Du Pont, that comes in 1802, when French emigre Eleuthere Irenee du Pont, a former apprentice to famed chemist Antoine Lavoisier, established a gunpowder factory on the banks of the Brandywine River near Wilmington, Delaware. The company flourished as a family owned and run firm, supplying both government and private sectors with gunpowder and, by the end of the nineteenth century, the newly developed technologies smokeless powder and dynamite.
After the Civil War, however, the fast-growing and increasingly competitive explosives industry was plagued by overcapacity and declining prices. Du Pont responded to this threat in typical nineteenth century fashion. It formed a cartel. Under the leadership of Henry du Pont, the Gunpowder Trade Association, more popularly known as the “Powder Trust,” was created in 1872 to control price and production. Representatives of the member companies met regularly to determine production quotas and prices and to arbitrate differences between or among members.
To help ensure compliance, Henry du Pont bought shares in most of the trust participants (i. e. , his major competitors). The strategy, which would not become expressly illegal until passage of the Sherman Antitrust Act in 1890, proved effective. By 1881, the association was estimated to control 85 percent of the U. S. gunpowder market. Du Pont pursued the same horizontal combination strategy for dynamite, with similarly successful results. The old trade-association model had worked well for Du Pont—better, in fact, than imilar arrangements in many other industries—but it was hardly perfect. There were compliance issues: Member companies sometimes kept two sets of books—one for the trade association or holding company, the other with the real numbers—and secret rebating, special discounts, and similar manipulations were rampant. What’s more, price and production agreements invited new entrants into the industry that, even if economically unviable, had to be bought into by the industry leaders if control was to be maintained.
Even when it was working smoothly, the old model had serious limitations. Under the association approach, there was little incentive to reduce costs, replace inefficient plants, streamline marketing and sourcing, or support research and development. The old guard had not worried about such issues. The new generation, less tied to the old ways – and well aware of the government’s growing legal scrutiny of trusts and cartels did.
Aided by talented managers like Moxham, du Pont cousins Alfred, Coleman, and Pierre set about remaking a chaotic nineteenth century, family-run trust—more a loose federation of small firms, really—into a unified, professionally managed, twentieth century industrial enterprise. They dissolved the Powder Trust, bought out the shareholders in many of its former members, and merged those firms, along with the ones that Du Pont already wholly owned, into a single integrated corporation.
In 1902, when the cousins had bought the firm, the du Pont company itself had manufactured about a third of the nation’s powder. By 1905, thanks to consolidation and centralization, the newly amalgamated firm’s market share had shot up to more than three-quarters. By 1906, Du Pont had dissolved some 64 corporations. To administer the newly consolidated enterprise, the cousins established a formal, centralized management structure with clearly defined lines of authority and responsibility.
The carefully crafted structure they adopted divided Du Pont into separate functional departments controlled by an Executive Committee (see Exhibit 1). This arrangement served the reborn company well, enabling Du Pont to exploit economies of scale to become the prohibitive leader in its industry over most of the next two decades. Thus, it remained basically unchanged until after World War I. In fact, Chandler notes, Du Pont’s centralized, functionally coordinated organizational blueprint proved “admirably suited” to meet the needs of the firm’s sudden, enormous growth resulting from the war in Europe.
At war’s end, the company’s structure seemed more ideal than ever. “By the summer of 1919, the Du Pont Company was ready to meet the postwar world,” Chandler writes. “Few more rationally planned and 2 thoroughly tested designs existed for coordinating, appraising, and planning the activities of a great vertically integrated industrial enterprise. ” But the centralized structure that had served Du Pont so well during one strategic era would turn out to be woefully inadequate for the next.
Diversification, World War I, and Reassessment Spurred by fears of becoming both ever-dependent on fickle government munitions contracts and too easy a target for federal antitrust actions, the company in 1908 began to investigate possible alternative uses for nitrocellulose, the highly flammable material that is the main ingredient in military gunpowder. Two years later, the firm made its first significant venture into the nonexplosives consumer market by purchasing the Fabrikoid Company, the country’s largest manufacturer of a nitrocellulose-based artificial leather.
This was followed by a tentative foray into the production of pyroxylin, another nitrocellulose compound that was used to make plastics products such as combs and eyeglass frames; lacquers; and photographic film. Even during the war, Du Pont anticipated the challenges it would face when the guns fell silent. “It is hoped,” the company stated in its 1915 annual report, “that new manufactures will be developed to take the place of the abnormal military business. ” Or as company Treasurer John J.
Raskob put it, “After the war it will be absolutely impossible for us to drop back to being a little company again; and to prevent it, we must look for opportunities, know them when we see them, and act with courage. ” Du Pont did aggressively search out and develop new capacities even as it was racing to meet the soaring wartime demand for munitions. This expansion extended both horizontally and vertically. In 1915, Du Pont bought the Arlington Company, a manufacturer of a brand of pyroxylin called Pyralin.
In 1916, it purchased the Fairfield Rubber Company, which produced Fabrikoid-based products such as rubber-coated automobile and carriage tops. That same year, it decided to enter the paint and varnish business (whose raw materials closely matched Du Pont’s existing capabilities), buying the large integrated paint firm Harrison Bros. & Co. In 1917, it bought the Marokene Company, which made a key ingredient in Fabrikoid. It also built a laboratory in New Jersey to study and manufacture synthetic dyes, a product whose chemistry was closely related to that of high explosives and a market that Du Pont had been exploring for several years.
Also in 1917, Du Pont invested $25 million to purchase 28% of the available stock in General Motors. Three years later, Pierre du Pont left the Du Pont company to become that firm’s president. 3 This brings us back to 1919, when this carefully planned, well-financed diversification should have been paying handsome dividends, but wasn’t; when the company asked a committee of junior executives to figure out why and when that committee concluded that the problem was the firm’s basic structure.
The committee began by rejecting the notion that Du Pont’s new ventures might be falling prey to macroeconomic forces, since its competitors (many of whom, unlike Du Pont, were actually making money) had “no advantage over us in the purchase of raw materials, and no secret process, no patents preventing us from using the best method of manufacture. ” As a result, the committee concluded, “the factor or factors making this great difference . . . are entirely within ourselves. In further studying the problem, the committee noted that although the firm’s functional departments were effectively managed, its organizational structure meant that there was no single person overseeing the operation of any particular product line. In Du Pont’s paint operation, for example, the committee reported that, “we have been unable to find the exact responsibility for profits”. So the committee turned its attention to Du Pont’s organizational framework—the unitary, or U-form structure that had been adopted in 1904 as the keystone of the newly consolidated, centralized enterprise.
That framework, the committee agreed, had served the company well over the past decade and a half. But— and this was their crucial insight—objectives change, and when they do, structures must adapt. The same formation that enabled a company to successfully execute one strategy might wind up hindering another. In other words, Du Pont’s existing structure—with its centralized manufacturing, sales, and purchasing departments that served all units—had worked well as long as those units were few in number and closely related. But that was no longer the case.
The essential difficulty was that diversification greatly increased the demands on the company’s administrative offices. Now the different departmental headquarters had to coordinate, appraise, and plan policies and procedures for plants, or sales offices, or purchasing agents, or technical laboratories in a number of quite different industries. The development of plans and the appraisal of activities were made harder because executives with experience primarily in explosives were making decisions about paints, varnishes, dyes, chemicals, and plastic products.
Coordination became more complicated because different products called for different types of standards, procedures, and policies. For although the technological and administrative needs of the new lines had many fundamental similarities, there were critical dissimilarities. In short, each product posed its own unique manufacturing and marketing challenges. The firm’s centralized, functionally departmentalized structure, however, presupposed and had been built to serve a much more homogeneous enterprises.
Restructuring Rejected and Realized For these reasons, the most sweeping recommendation of the committee’s March, 1920, final report was that Du Pont overhaul its basic organizational structure. The firm ought to replace most of its functional departments with a pair of separate, largely self-contained divisions based on product lines: one for paints and chemicals, the other for plastics and celluloid articles. Each of these units would function semiautonomously, with its own managing director overseeing proprietary purchasing, accounting, manufacturing, and sales departments.
They would be supported by centralized departments like Development, Engineering, and Treasurer’s, and a single, top-level Executive Committee would continue to chart the corporation’s overall course. But the responsibility for day-to-day operations and decision-making would reside squarely within the product divisions (see Exhibit 3). At the time, however, it was a radical notion. In both theory and practice, the U-form was the undisputed organizational standard of its era. Though some American railroads, most notably the Pennsylvania, had in the 1880s evised “line-and-staff” taxonomies that combined departmental (“line”) and central office (“staff”) structures, American industry offered no precedents for reorganizing around product divisions. It was, in fact, more innovation than Du Pont was prepared handle. The restructuring proposal met with strong skepticism from top-level management, many of whom remained unconvinced that the company’s woes reflected anything more than the inevitable problems that attend the pursuit of new markets. The answer was not reorganization, they said, but better information and knowledge.
It was not the development of new offices and new lines of authority and responsibility but rather the fashioning of more effective inventory controls, or more accurate volumes, sales, and market figures, and of other data to flow through the existing lines of communication. Such information would help prevent unnecessary purchasing and make it possible to find just where the high manufacturing and selling expenses were located. Thus, senior management questioned the committee’s assertion that underlying structural issues were to blame.
How did the committee know this for certain? How could they be confident that an organization-chart overhaul would fix the problem? Why instigate revolution when reform might be sufficient? At first, conservatism won out. In November, 1920, despite the fact that yet another committee analyzing Du Pont’s internal organization had echoed its predecessor’s call for a new, product-based arrangement, President Irenee du Pont vetoed the restructuring proposal. The groundbreaking plan seemed dead. But two things conspired to keep it alive.
First, a separate, unofficial working group examining the company’s poor performance in the paint business made a series of recommendations that, while not explicitly focused on the question of organization, did assert the need for a product-specific management approach in that area. Running the paint business more efficiently, the group concluded, would require that “the responsibility of profits and the control of the business be in [the] same place”—a Paint Steering Committee whose chairman would be the de facto managing director of the entire product line.
This proposal was adopted, and when it proved fruitful, similar “divisional councils” were created to oversee other product lines. Thus, though the formal, companywide restructuring plan was officially off the table, more and more of its underlying logic was permeating the firm. The second phenomenon helping to resuscitate the restructuring proposal was the business cycle— specifically a brief but sharp depression that rocked the U. S. economy from 1920–1922 and created crisis conditions at Du Pont. In the first six months of 1921, the company’s diversified products lost more than $3. 8 million. Explosives, however, continued to be profitable). For the full year, the firm’s net income was about half of what it had been in 1920. It laid off two-thirds of its hourly workers and half of its salaried employees. By the summer of 1921, it was increasingly clear to Du Pont’s top management that incremental changes and a conservative approach would not suffice to pull the firm out of its tailspin. Drastic measures were needed. In August, amidst growing fears about the company’s long-term prospects, Du Pont’s senior executives met to resurrect and reconsider the reorganization proposal.
In September, the Board of Directors approved a new decentralized administrative structure centered around products rather than functions. In October, the transformation began. The new structure approved by the Board (see Exhibit 5) embodied the same product-centered, line-and-staff principle as the junior executives’ original proposal from a year and a half before (see Exhibit 3). But it had grown in scope and complexity. The scheme divided Du Pont’s manufacturing operations into five product departments—Explosives, Dyestuffs, Pyralin (plastics), Paints and Chemicals, and Fabrikoid and Film.
Atop each was a general manager with day-to-day responsibility for purchasing, production, and sales in that department and accountability to the Executive Committee for the department’s performance. General company support services such as legal, research and development, personnel, and advertising were dubbed “auxiliary departments” and remained central operations. Key to the plan was its separation of operational responsibilities from overall strategic planning and coordination. No general manager of a department would serve on the Executive Committee.
In the words of H. Fletcher Brown, manager of the Smokeless Powder Department and one of the chief architects of the new organization, this would free the department head to “give his undivided attention to the details of his own Department,” while allowing the Executive Committee members to “give all their time and effort to the business of the Company as a whole. Being connected with no department, they will be able to consider all questions or problems without bias or prejudice. The new strategy of diversification and the new structure of autonomous product divisions meant that top professional managers no longer headed a plant or a functional department. Instead, they assumed direct responsibility for key decisions about resource allocation and organization in a general office. There they systematically coordinated and appraised operations in all Du Pont lines while planning for expansion and seeking new opportunities for investment. Postscript The multidivisional or “M-form” structure that Du Pont pioneered in 1921 would go on to become the standard organizational model for the modern diversified corporation.
According to business historian Charles W. Cheape, “The innovation of the decentralized structure organized around product divisions was a major watershed in the history of Du Pont and of American business in general. ” It allowed “the economies of scope that made the large, diversified firm possible. ” Economist Oliver Williamson went a step further, calling the multidivisional form the most important innovation of capitalism in the twentieth century. Alfred D. Chandler, Jr. , argued that virtually every successful company in the U. S. etween 1920 and 1960 adopted the M-form, and that it played a key role in the rapid growth of American businesses through this period. And the broader lesson that Chandler drew in part from his careful study of Du Pont—that a firm’s structure should reflect, and, if necessary, change with its strategy—has become a staple of management theory. As for the impact of the reorganization on Du Pont itself, it is difficult to pin a particular cause to a precise effect. With so many factors, both internal and external, at play, who can know for certain the impact of one decision on the performance of a company?
Nevertheless, Du Pont’s adoption of the M-form structure seems to have achieved the desired results. Company performance began to improve in 1922 and continued to strengthen throughout the decade—although at least some of that upswing reflected brightening macroeconomic conditions (see Exhibit 7). As early as 1921, the firm was reporting that although “sufficient time has not elapsed for the expected effects to be fully realized . . . .it seems very clear that important advantages have been gained. ” By 1922, it could declare unequivocally, “This type of organization is very definitely yielding good results.
