Unmanageable Risks: MacPherson v. Buick and the Emergence of a Mass Consumer Market

On May 17, 1910, Donald C. MacPherson purchased a Buick runabout from the Close Brothers dealership of Schenectady, New York.[1] The new rig sported a “four cylinder, twenty-two and a half horse power” engine, allowing it to reach a speed of fifty miles per hour. Its body had been painted “French gray” and a similar gray color coated the wooden wheels. The runabout accommodated two passengers in the front seat and one in the rumble seat. Though smaller and slower than other cars on the market in 1910, the Buick served MacPherson’s purposes. During the summer and fall, he drove the machine to various places in the vicinity of Saratoga Springs for his business as a stone cutter who specialized in making grave stones. Like many motorists in this early market, he stored his runabout in a barn that winter and put it back into service the following May. The auto gave him no serious trouble until July 25. While traveling at a moderate speed along a road leading into Saratoga Springs, the car’s left rear wheel collapsed; the machine overturned and trapped MacPherson beneath the rear axle.[2] He later testified that his “eye (right eye) [had] torn apart entirely, [and] laid down from the eye brow,” and that his right arm “had broken at the wrist.” He suffered such great pain in his right wrist that for many months he could not perform his work as a stone cutter as he lacked the strength to grip his tools. His eyes deteriorated: First his right eye and then his left eye began failing, making it difficult for him to identify clients and friends.[3]

Two months after his accident, MacPherson turned his bad experience into a lawsuit. Complaining that the hickory spokes used for the car’s left rear wheel were rotten and that this defect had caused the wheel’s collapse, he sued the Buick Motor Company (a division of General Motors) for negligence. The first trial ended when the judge dismissed the complaint. MacPherson appealed, and the state’s Appellate Division for the Third Department reversed the judgment, granting a new trial. In March 1913, the jury rendered a verdict for MacPherson amounting to $5,025.[4] Buick then appealed, but in January 1914 the Appellate Division affirmed the judgment for MacPherson. Justice John M. Kellogg found that Buick “owed a duty to all purchasers of automobiles to make a reasonable inspection and test to ascertain whether the wheels purchased and put in use by it were reasonably fit for the purposes for which it used them.”[5] Buick appealed again. In March 1916, speaking for the Court of Appeals of New York, Judge Benjamin Cardozo concluded that Buick “was not at liberty to put the finished product on the market without subjecting the component parts to ordinary and simple tests.”[6]

In MacPherson v. Buick (N.Y. 1916), economic and legal questions overlapped. In economic terms, the plaintiff demanded to know who, for the sake of innovation, should bear the costs of an imperfect technology in a new market. In legal terms, Cardozo took up the question of how the line should be drawn between dangerous and nondangerous goods and what responsibility corporations owed consumers for a product’s quality.[7] A critical part of the case concerned privity of contract, the legal requirement that the plaintiff have a contract with the party being sued. MacPherson lacked this contractual relationship with Buick because he had purchased his car from the Close Brothers dealership, and the franchised car dealer had bought the car from Buick in the fall of 1909.[8] Buick thus maintained that the dealer, not the manufacturer, was liable for the defect. Rejecting this argument, Cardozo put aside the legal barrier Buick had achieved through the franchise contract and held the manufacturer accountable for MacPherson’s injuries. Before MacPherson, the courts had created exceptions to the privity requirement for inherently dangerous goods, such as poisons.[9] Cardozo sought to restrict the privity doctrine directly: “If to the element of danger there is added knowledge that the thing will be used by persons other than the purchaser, and used without new tests, then, irrespective of the contract, the manufacturer of this thing of danger is under a duty to make it carefully.”[10]

This article assesses the relationship between the changes in liability rules as related to defects and the emergence of a mass market for automobiles. I begin with the premise that social costs were inherent in the process of creating a new market.[11] Rather than wait till they had perfected the technology before selling machines, automakers sold imperfect vehicles. In some cases, the defect was as simple as a spark plug that failed to spark. In others, as serious as an axle that broke because it had been improperly heat treated. Henry Ford built his first auto in 1896, and inventors, entrepreneurs, and corporations spent the next twenty years solving many technological problems.[12] As automakers worked to improve their products, they sold defective cars, like MacPherson’s runabout.[13]

Corporations thus faced a choice between acquiring the knowledge needed to design a safe product and exercising power in the market to impose the costs of defects on consumers. This choice forms my narrative. In Part I, I describe the relationship between a new technology and consumers’ risks of physical injury in the early auto market. In Part II, I find that managers took advantage of the privity requirement in structuring the modern firm. The dealer franchise contract initiated long-term relations between manufacturers and retailers so as to foster the market’s growth, but also to shift many costs of innovation, including the costs of defective vehicles, to dealers or consumers. Part III treats MacPherson’s lawsuit. Before Cardozo’s ruling, Ford managers had begun inspections and used them to defend the firm against consumers’ suits. By comparison, MacPherson was fortunate first because Buick had conducted no tests of its wheel in 1909, and second, because he located a source of expertise outside of Buick’s control and used it to demonstrate the defect. Part IV assesses the impact of Cardozo’s ruling. Automakers reduced defects, but the ruling alone did not cause this decline. Managers improved their vehicles’ quality to win consumers’ repeat purchases in a mass market. In addition, many oversight agencies instructed managers to better their vehicles’ quality for their own goal of selling cars to consumers. Part V traces the impact of stiffer safety standards in terms of managers’ new demand for products liability insurance.

My analysis of the automobile market calls attention to the impact of liability rules not only on the costs of innovation, but also on the structure of the modern firm. Manufacturers relied on the privity requirement to saddle car buyers with the costs of defects. I argue that this liability for defective vehicles conditioned managers’ decision to use the franchise sales contract as their method of mass distribution. It is second nature for scholars to speak of production costs and transaction costs, but they rarely have considered liability as a cost of doing business. This article examines the ramifications of liability for defects for the structure of the modern firm.[14] As the market evolved, manufacturers formed long-term ties with market actors, including their dealers, as Stewart Macaulay explained. Yet, they still counted on those contracts to deny these relations in court for the sake of avoiding liability. Writing before Macaulay, Friedrich Kessler studied the power corporations exercised through the franchise sales contract rather than an agency relationship. I offer an explanation for their choice of the franchise contract in the particular context of the early auto market.[15]

MacPherson intended to stop corporations from avoiding products liability. Legal scholars have compared it to the many cases that led up to Cardozo’s ruling, as well as to those that followed. Steven P. Croley and Jon D. Hanson, for instance, identified MacPherson as the first of four landmark products liability cases in the twentieth century. Their analysis indicates that MacPherson represented a break with the past, but how sharp a break and in what direction depended in part on subsequent cases. Its impact also depended on other factors that impinged on managers’ approach to product quality, notably the market. Finding fault with recent legal writings, Croley and Hanson singled out “market failures” as seen in terms of imperfect information, manufacturers’ power over consumers, and the question of distributing risks.[16] I examinine market problems in terms of the evolution of one product. Before Cardozo’s ruling, manufacturers had begun to test vehicles. During the 1920s, they made more sustained investments in R&D;, wanting to improve their vehicles’ reliability, durability, and safety in their effort to secure consumers’ repeat purchases in a mass market. But there is another conclusion to be drawn from this analysis. Had market pressures alone ensured a safe product, there would have been no need for other sources of discipline. Manufacturers acknowledged MacPherson when they improved their methods of inspection. More importantly, a mix of private and public organizations, including engineering societies, federal research entities, insurance underwriters, and state regulators, established safety standards for critical components such as safety glass and lighting systems. Managers absorbed (to varying degrees) the regulatory lessons about the importance of product quality in shaping their efforts to secure consumers’ loyalty. Put another way, the regulatory oversight gave force to Cardozo’s ruling for specific components while never completely ensuring a vehicle’s sound construction.

My study of early automobiles focuses on themes familiar to legal historians: corporate power, technological defects, and the role of the state.[17] Depending on the social context and the technology in question, historians have characterized the relationship between defects and injuries in different ways. In her study of railroad accidents, Barbara Young Welke argued that railroad companies and the courts maintained that male passengers could take risks not taken by women and compensated women to a greater degree than their male counterparts.[18] Looking at workplace injuries, Arthur F. McEvoy and Christopher L. Tomlins argued that management’s ability to increase the speed of machinery or otherwise reorganize technology put laborers at greater risk of injury.[19] In the case of a new complex consumer product, I maintain that management’s inability to control the technology combined with their ability to innovate in a market context meant that firms sold defective products and consumers became risk-takers.

Existing studies about market innovation, however, have paid little attention to consumers’ risks of injury or the impact of liability rules on a new market’s development.[20] Indeed, although the literature about automobiles is vast, scholars have overlooked liability for defects in terms of who bore the costs of innovation.[21] This omission, I argue, both reflects a particular view of innovation and invites a rethinking of the concept. The literature about innovation follows in the path of the economist Joseph Schumpeter, who singled out the role of entrepreneurs—unusual individuals who had the capacity to see beyond daily routines, fight entrenched interests, and introduce new ideas.[22] Economic scholars, of course, have modified Schumpeter’s ideas, but it is still second nature for them to treat innovation as an activity undertaken on the market’s supply side by inventors, entrepreneurs, managers, mechanics, and other individuals who worked inside firms.[23] Without disputing the fact that actors on the market’s supply side faced risks, I argue that car buyers (consumers and dealers) also shouldered risks in the creation of a new market.[24] According to my premise, actors on the market’s demand side incurred social costs in the form of financial losses and physical injuries as an inherent part of the process of innovation. Recognizing their risks, I trace the impact of liability for defects on the modern firm’s structure and corporate research.[25]

I rely heavily on corporate records, especially those from Arthur D. Little, Inc. (a consulting and research company), General Motors, and the Ford Motor Company, as well as the records and briefs of many court cases. These documents reveal business practices often obscured by the privity requirement. In the case of industrial injuries, McEvoy and Tomlins argued that the nineteenth-century labor contract worked to deny men and women compensation for workplace injuries and, by extension, to hide the business activities that had produced the accidents.[26] The legal requirement of privity of contract had similar effects. As courts privileged this contractual issue, they denied consumers compensation and left unquestioned the ways in which corporations had designed the vehicles causing the accidents. Recovering corporate records enables me to examine the impact of liability rules on corporations and the market.

I. Risks of Innovation, Risks of Injury

In 1896, Charles E. Duryea and his brother, J. Frank Duryea, who were credited for constructing the first viable gasoline-powered automobile, sold their first vehicle; estimates place total sales at fifteen that year. In 1900, automakers sold roughly 4,000 vehicles; ten years later they sold 187,000 machines, including MacPherson’s runabout. In 1916, two decades after the market’s start, sales reached 1.6 million.[27] As the market increased in size, the construction of vehicles changed dramatically. At least until 1904, vehicles were small and crude. The Merry Oldsmobile, powered by a one-cylinder engine, claimed a third of the market in 1903. Between 1904 and 1908, manufacturers began producing touring vehicles equipped with four-cylinder engines and capable of seating four passengers.[28] In 1908, Henry Ford introduced his Model T, and Billy Durant organized General Motors. In 1912 Cadillac made standard Charles Kettering’s electric self-starter—a device that eliminated hand cranking, thereby making cars easier to start and encouraging buyers to enter the market. By 1913, Ford’s imaginative mechanics had an assembly line working at Highland Park.[29] Between 1896 and 1916, inventors and entrepreneurs achieved considerable improvements in this complex technology, yet innovation in a market context meant that firms initially sold highly imperfect machines. Car buyers thus faced two options: They could absorb the costs of defects or demand compensation. In the market’s first ten years, many motorists accepted the risks embodied in crude machines. With time, automakers constructed more reliable machines and fostered a commercial market, but their vehicles still housed defects. Corporate documents and court records from 1904 to 1908 reveal the transition many car buyers made from patience to exasperation to lawsuits.

At least until 1904, the automobile’s rudimentary character created many technological defects. In 1901, writing for Horseless Age, the industry’s leading trade journal, Albert Clough complained about axles. “If a manufacturer finds that the axles of his machine are giving way [breaking] the next lot of vehicles are provided with axles of a slightly larger diameter and so on until they begin to stand up pretty well.” This approach ultimately yielded improvements, but he added, “it is in no sense mechanical engineering and is fraught with danger and dissatisfaction.”[30] In 1902, the journal’s accident column reported that an Ohio driver had been pitched from his machine when an axle broke.[31] Rudimentary brakes presented another risk. One motorist told of his brakes being made with leather straps so that, as the brakes became hot, the leather “carbonized.”[32] Henry Ford’s biographer recalled that the 1904 Model A’s brakes “were good if assembled with great care,” but that the rush to assemble resulted in defects.[33] Crude steering devices posed another risk. In Valparaiso, Indiana, a chauffeur was seriously injured and his passenger killed when “[a] screw, it was said, came loose and control of the machine was lost.”[34] Fires resulted from exposed electrical wires; leaks in gas tanks and carburetors; and backfiring in carburetors, attributable to a sticky valve or an improper mixture of gas and air.[35]

The terms of insurance policies warned car buyers about their expensive products’ poor quality. In the 1890s all an owner could obtain was fire insurance, and this coverage was limited to a car’s storage—”when at rest in the owner’s stable.”[36] Car owners were required to fill gas tanks in “daylight only” and not to allow “a blaze or artificial light” in the room when filling the tank.[37] In 1902 auto fires prompted Boston underwriters to raise rates, as Horseless Age reported: “rates are high everywhere, and yet the companies are not at all anxious to secure the business.”[38] In 1903 The Automobile noted that underwriters “do not find [liability coverage] a profitable class of business, and a large number of them have, from time to time, withdrawn from the field, leaving the demand much greater than the supply.”[39] Not until 1906 did underwriters offer coverage for “damage sustained by the machine itself, as well as against any liability of the owner for damage to other property arising from the collision.”[40] Even in 1908, for coverage of liability and property damage, Fidelity & Casualty, Pennsylvania Casualty, and United States Casualty, as well as other underwriters, were “leaving it strictly alone.”[41]

Despite this state of affairs, some consumers purchased automobiles. The buyers featured often in the press were men and women from New York’s society pages, who could hire chauffeurs or buy new cars each year.[42] But also typical was H. G. Buschman, who declared to an Olds dealer in 1903 the desire to be the first motorist in Prescott, Wisconsin.[43] The vast majority of motorists were male, and the trade press celebrated a sense of male risk-taking, replaying stories of men who achieved great feats—new hills climbed, distances covered, and mechanical puzzles solved.[44] The drivers were rugged. Some were also daredevils who, declared Horseless Age, had more money than brains and drove too fast for the safety of pedestrians, other vehicles, and their own bodies.[45] Above all, car owners celebrated their mechanical know-how. In 1902, an anonymous motorist reported a near mishap. Out for a ride, the car’s steering wheel suddenly could “turn freely upon the column without doing any steering.” Had the vehicle been traveling at a high speed, the writer felt sure a horrible accident would have ensued. Yet rather than threaten a lawsuit, the owner explained how to fix this machine. Because “the threads in the aluminum wheel had stripped,” the steering gear came loose. “I put a good, honest steel key into the wheel hub and column,” the driver reported, “and it has not yet become corrupted by the company it keeps.”[46] Other car buyers expected to fix vehicles themselves. In March 1903, C. H. Farnum of Baraboo, Wisconsin, wrote to Ransom Olds’s brother, E. W. Olds, a car dealer, and complained about his carburetor, asking Olds to “give me what information you can.” To another car owner with carburetor problems, Olds wrote, “the gasoline mixture is a very delicate thing, a little rich, and the mixture will cause loss of power and waste of gasoline, so that a quarter of an inch turn on your needle valve means a good deal.”[47]

In the ten years after 1905, manufacturers began producing sturdier vehicles and establishing business practices necessary for a commercial market. The automobile historian James Flink concluded that “the 1908 state-of-the-art gasoline automobile was a fairly reliable family car.”[48] By 1913, Henry Ford’s moving assembly line had gone into operation. To make mass production possible, parts needed to be interchangeable, and by making parts uniform, Ford’s vehicles reduced defects produced by deviations from a set standard.[49] This method of production, of course, did not stop defects inherent in the design of parts. Although companies in a few industries, notably railroads, had established research laboratories in the late nineteenth century, most firms in other industries maintained small testing labs to assess materials on their arrival to a plant.[50] In the auto industry, Horseless Age noted in 1907 both the need for and lack of adequate research facilities.[51] Within a few years, however, public and private entities had begun to establish research institutions. The year 1910 ushered in two noteworthy research organizations for automakers. First, the Society of Automobile Engineers (SAE), organized in 1905, established standardization committees to investigate components and materials and recommend codes for uniform practice.[52] A second major change for automotive research came in 1910 when the U.S. Department of Agriculture (USDA) established the Forest Products Laboratory (FPL) and assigned it the task of undertaking research projects to conserve wood and reduce waste.[53] Just as the FPL and the SAE pursued many projects, individual automakers began funding research, and their staffs helped make autos “fairly reliable.” But the engineers also identified many problems yet to be solved and thus acknowledged many persistent defects.

In 1911, General Motors hired Arthur D. Little, Inc., a consulting company that performed research for many companies.[54] Claude E. Cox ran the consulting laboratory for GM in Detroit. His staff included researchers in chemistry, metallurgy, electrical and mechanical engineering, paints, and materials. They tested components (wheels, tires, magnetos, carburetors, brake linings, valves, piston rings, batteries) as well as materials (oils, steels, paints). Their reports sought to identify cheaper and more effective production methods as well as stronger and more reliable materials. Their watchword was standardization, as they wanted improved practices adopted throughout the GM divisions.[55]

Striving to better the design and production of vehicles meant that consultants called attention to existing problems. One of the first reports, for instance, concerned the strength of wooden wheels. The test was conducted “in connection with an action for damages brought against the Olds Company alleging improper design of a front wheel of an Oldsmobile.” (Like Buick, Olds by this date was a division of General Motors.) In March, at the Imperial Wheel Company, where Buick purchased its wheels, the consultants tested a 36-by-5-inch wheel. “A total stress of one ton caused a cracking sound, and was evidently above the safe limit.” “It would seem,” the authors continued, “that the hub design could be strengthened with advantage at very small additional expense.” They added in their conclusion, “We think careful tests of a number of wheels of various types should be made with a view to developing a wheel of substantially equal strength at all points.”[56]

Even if components were properly manufactured, still they could put so much stress on another part as to cause it to weaken and break. In 1911, Cox complained about the clutch’s “violent action.” If poorly designed, its sudden engagement exerted pressure on other parts of the vehicle, such as the rear axle. Repeated jarring initiated by the clutch could cause the axle to harden, or become brittle and break.[57] As cars became heavier and faster, their added weight and speed also put parts at risk of breaking. For instance, in 1909 Horseless Age reported an increase in accidents from broken wheels. The diameter of wheels had increased, but there was no increase in the diameter of the hub or the width of the spokes. The relative changes made the wheels weaker and less resistant to lateral strains.[58] Poorly designed parts also posed fire hazards. In 1911, Cox reported that for the Oldsmobile roughly “12% of all cars shipped caught fire from back-firing through the carburetor.”[59]

Cox’s researchers further complained about the improper use of materials. In June 1911, they reported “a defective shaft” made with “very low carbon steel instead of 3–1/2% nickel steel as specified.” In November 1911, they traced “brittle nickel steel” to the “poor practice at the rolling mills.” That December, Cox further complained that “steering arms have broken, due to a low factor of safety and low grade material not properly heat treated.” His staff, in turn, called for the “systematic and thorough inspection of such incoming material before acceptance.”[60]

At GM, Claude Cox grew frustrated with the firm’s research standards. He claimed in 1911 that other firms devoted more money to experimental research than “the entire amount spent so far on the Research Department” and that some GM divisions were not “equipped for experiment work or research work of any kind.” In 1912 Cox fumed about the lack of coordination between his lab and specific companies.[61] In early 1913, GM cut back its work with Arthur D. Little, and the records suggest the relationship ended that year. In their three-year affiliation, the consultants had assisted the automaker in many regards such as in their efforts to introduce uniform policies. Yet, the relationship also exasperated Cox, as GM appeared to be too strapped for cash or too disorganized to fund the research on the scale that he expected.[62]

Just as Claude Cox faulted GM’s automobiles, motorists found that their purchases did not live up to their expectations for a reasonable commercial transaction. Some consumers continued to repair their vehicles. Others expected automobile dealers to fix problems, but not all problems could be remedied easily.[63] The most serious were personal injuries, but consumers voiced more frequent complaints in trying to operate their machines. W. Benton Crisp complained to Ford’s New York branch manager that, within two months of ownership, his 1905 car had required two new axles. He heard that Ford had had so much trouble with its supplier that officials “had either commenced or were about to commence an action against” the supplier “for breach of contract in supplying the faulty axles.” Crisp concluded: “if the Ford Company [was] damaged by that transaction, each of its customers must likewise have been damaged.”[64] Morton E. Duncan was sufficiently exasperated with his Model B that in 1905 he sued Ford on account of the car’s “defective construction.” His settlement required that Ford install new “cooling apparatus, new steering column, steering wheel and new dash-board.” Provided Ford carried out these requests, Duncan agreed to “use his best influences to promote the sale of Ford Motor Company’s cars in the City of Philadelphia and vicinity.”[65]

Other car buyers were not as charitable as Duncan. Many lawsuits finding their way to the appellate courts concerned breach of warranty and efforts to recover the full sum buyers had paid for their vehicles. In Beecroft v. Van Schaick (N.Y. 1907), Edgar Beecroft had purchased a car with an express warranty but found that despite “repeated attempts to remedy the defects” the car failed to perform satisfactorily. He sued for the return of $685 that he had originally paid for the vehicle.[66] After a car owner in Wisconsin attempted several times to have the dealer make the car operate properly, he also sued to recover the purchase price.[67] In some cases, the seller won the dispute. In 1906 Morris Osburn purchased a vehicle from Ford’s Chicago branch. At the trial, the municipal judge observed: “the machine when received by the plaintiff, and subsequently thereto, did not operate properly, and we do not understand this to be denied by the defendant.” The judge added, “nothing seemed radically wrong with it” and thought a good mechanic could put the car in proper order. Yet, the good mechanic did not fix the machine, and instead Osburn shipped the vehicle from Wisconsin back to Chicago. The car burned in a fire, and its destruction set the stage for the lawsuit. Ford, however, could have avoided this conflict had the company manufactured a vehicle free of serious defects.[68]

Defects figured as well in conflicts between dealers and manufacturers. Arthur F. Neale sued the American Electric Vehicle Company (EVC) in 1902 complaining that he had been unable to sell a vehicle to the Jordan, Marsh Company because it was defective. He further contended that the loss of this sale so tainted his products he could not sell any EVC vehicles in the local Boston market.[69] In 1905, two dealers similarly complained that the Wayne Automobile Company sold them such defective vehicles they failed to make nearly as many sales as they thought they should have.[70] In 1904, the Reid Manufacturing Company refused to pay its supplier, Buick Motor Company, claiming the parts were defective. Reid managers pointed to costly repairs their Oakland dealer undertook as well as sales they lost to their New York agent.[71] In 1909, a Georgia dealer canceled his 1908 contract, explaining to the manufacturer: “several White Steamers belonging to Atlanta owners have been badly damaged or destroied [sic] by fire during the past few years, which fact seems to have created considerable local prejudice against your car.”[72]

Risks that a car would catch fire, skid, not run, not stop, or stop abruptly were traced to the design of components, materials used to manufacture parts, the assembly of parts, and the relationship of parts to the machine’s overall strength. In the market’s first ten years, the vehicles’ rudimentary character had advertised many defects to car buyers. Insurance policies’ limited coverage further warned car buyers to be wary of crude machines. As the market grew in size, machines became reasonably reliable. But while safer in some regards, many still housed serious defects. Testifying in MacPherson’s trial, Alanson P. Brush, a GM engineer, was asked how he determined a part’s durability. “The most satisfactory information that we can have, in fact the only means to the designer,” he stated, “is to use the customers, that is to go over the complaint correspondence. That is the most satisfactory information a designer can have.”[73]

II. The Privity Requirement and the Structure of the Modern Firm

In 1913 J. M. Livesay sued the Ford Motor Company, claiming that a defective wheel had caused his accident and injuries. The court ruled for Ford, because the plaintiff had bought the car from a Ford dealer and thus lacked a contractual relationship with the manufacturer.[74] During much of the nineteenth century, consumers lacked a contract upon which to sue manufacturers simply because most corporations were not integrated. Companies typically distributed their wares through wholesalers, jobbers, or other middlemen who peddled their goods to local retailers—grocers, druggists, dry goods merchants, who in turn sold products to consumers.[75] This relationship changed when, for greater efficiency, managers integrated distribution within the modern firm. Retailers became a firm’s sales agents and the firm became liable for their sales agents’ actions. Automakers, by contrast, placed concerns about liability for defects ahead of gains in efficiency when they arrived at a method of distribution as a system of franchises.

As Alfred Chandler explained in The Visible Hand, the integration of mass distribution and mass production was distinctive to the modern firm. They constituted critical links in a chain of business activities from the acquisition of raw materials to a good’s final sale, a process Chandler called “vertical integration.” Precisely because of their products’ technological complexity, managers wanted to integrate distribution within the firm. That is, they wanted specialized sales agents to market and service their products. Examples included Remington, a maker of office machinery, and manufacturers of heavy machinery such as the Otis Elevator Company, Western Electric, and Babcock & Wilcox.[76] Charles McCurdy similarly noted that the sewing machine manufacturer, Singer, replaced independent retailers with its own retail outlets by 1879.[77] Between the 1880s and the early 1900s, many corporations established marketing systems based on salaried sales agents. Although Chandler did not explore the legal ramifications of these decisions, firms that relied on salaried agents were liable for their agents’ actions.[78]

Not all firms fit this pattern, however. When automakers defined their relations with dealers through a franchise sales agreement, Friedrich Kessler called this method of distribution “vertical integration by contract.” He found that by the 1950s franchising applied to many complex goods—autos, tractors, farm implements, tires, and electronics.[79] Auto dealers, much like the retailers Chandler cited, sold a complex machine and sunk large investments in their local markets. Although manufacturers did not make dealers salaried employees, they wanted on-going relations with their retailers. They chose between two different arrangements—an agency contract or a franchise sales contract. The courts ruled that where cars were sold outright to a dealer (not on commission) and where a dealer was prohibited from making a contract with car buyers in the company’s name, the contract was treated as a sales (not an agency) contract. The distinction mattered to a firm’s administrative control and its liability. With an agency contract, the manufacturer, as the principal, directly controlled its dealers, as its agents. Further, the manufacturer as principal was liable for its dealers’ actions. A sales contract, however, implied an immediate exchange: Firms sold cars to dealers at a discounted price and dealers resold cars to consumers at a retail price. As such, manufacturers held no direct control over dealers. They also were not liable for their retailers’ actions and not liable to consumers as third parties.[80]

When E. Wells Johnson sued Cadillac for negligence in the sale of a motor car whose defective wheel caused his accident and injuries, Cadillac relied on its 1908 dealer contract to claim that the car buyer lacked a contractual relationship upon which to bring his lawsuit.[81] Cadillac’s experience poses two questions. First, was the company typical of other leading manufacturers—that is, had many firms adopted the sales contract by 1908 or 1909? Second, had the managers of Cadillac and other automakers, such as Ford and Buick, registered the importance of liability for defective vehicles by 1908 or 1909? My answers are framed in part by a collection of dealer agreements, described in the Appendix, and in part by my reading of court cases.[82] When the industry was new and many firms were just being organized, managers wrote short contracts with few provisions. Ford and other companies also wrote a variety of contracts, including agency contracts with salaried managers.[83] Cadillac’s sales manager noted in 1902 that the firm had “no regular contract.”[84] Yet, by 1908 or 1909, many companies had clarified each party’s obligations and limited their liability. Most important, contracts declared that the dealer was not the firm’s agent. Ford added this clause in 1904. By 1908, the list included leading producers, such as Cadillac, Jeffery, Oakland, and Reo.[85]

Managers may have initially chosen to rely on the sales contract for reasons unrelated to defects or even unrelated to liability in general. Allan Nevins, for instance, reasoned that the franchise better motivated dealers to sell cars than did a flat salary.[86] Legal scholars, by contrast, called attention to different types of liability that had little relation to defects. The Deere-Clark Motor Car Company, for instance, began making cars in 1906. But after a labor strike stopped production, lawsuits were filed for failure to deliver cars, and the firm declared bankruptcy in 1907. Suppliers’ delays in shipping parts also threatened to halt production and delay shipment of cars from the factory to dealers.[87] By 1909, several companies had adopted clauses exempting themselves from any costs resulting from delays due to unforeseen events, typically strikes or fires.[88] Still, these factors need not have been mutually exclusive. Managers may have worried about their liability for delays in shipping cars at the same time they tried to limit their liability in selling defective products. What is important in my account is that soon after firms had been organized, managers quickly learned about the consequences of selling defective products. They did not necessarily worry about liability for personal injury, but they were sued by dealers and consumers for reasons that at root were prompted by their machines’ defects. Ford, Cadillac, Buick, Reid, EVC, Wayne, and Pope all faced lawsuits prompted by defects soon after their organization.[89] They could limit their liability through agency and sales agreements, but the two options presented tradeoffs as firms discovered in a set of cases decided between 1904 and 1908, as well as other cases initiated during these years.

In two early cases, managers tried to defend themselves by claiming their dealers were their agents. In Neale v. American Electric Vehicle Company (Mass. 1904), Arthur F. Neale sued EVC, complaining that its vehicles never lived up to the warranty claimed to exist in the contract. The manufacturer first denied that the contract contained a warranty and second declared that it owed no responsibility for making the machines “merchantable” to the dealer as its agent. Its lawyers reported that the dealer was “paid a commission of 20 per cent of the purchase price.” Further, they wrote, “It is a novel legal proposition that a principal in a contract of agency for the sale of chattels is under a duty to his agent, implied by law, to have the chattels merchantable…. If no such duty is owed, it is plain that the defendant is not liable for not performing it.”[90] By contrast, in Masters v. Wayne Automobile Company (Mass. 1908), the plaintiff’s brief drew the careful distinction between an agent and a vendee: “the Court holds that a man may be an agent as to some things, and, at the same time, an individual contractor as to others.” Where the dealer acted as a vendee in this instance, “the defendant company is liable upon its implied warranty that its automobiles were of merchantable quality. The Master finds and cites certain instances where said cars as delivered by the defendant company were defective and that some would not run without fixing nor climb hills as other automobiles ordinarily do.”[91] The attorneys thus persuaded the court to accept their interpretation. A similar distinction applied in Wheaton v. Cadillac Automobile Company (Mich. 1906). The dealer sued the manufacturer for failure to deliver goods, but Cadillac succeeded in claiming an agency relationship for which it was not liable to its agent for not having shipped the cars.[92]

Although an agency relationship offered manufacturers a defense against cases rooted in the twin problems of defects and delays, it also left automakers open to other types of lawsuits. Ford’s experience was instructive. While the company wrote a sales contract with dealers in 1904, officials also wrote contracts with branch managers in which the manager was a salaried employee. Thomas Hay held one such contract as Ford’s Chicago branch manager, and it was upon this contract that Morris Osburn sued Ford instead of Hay. Although Ford ultimately prevailed in the 1908 case, managers could have avoided the lawsuit altogether had the branch manager not been the firm’s agent.[93] George Joslyn gave Cadillac managers a similar lesson when he sued the company for product misrepresentation. Joslyn did not charge that the dealer had made false statements about the vehicle, but instead maintained that the car was defective. In court, Cadillac tried without luck to shift liability to its sales agent, arguing that the sales manager was in fact a dealer like any other dealer.[94]

The sales contract allowed managers both to sidestep the lawsuits with disgruntled consumers and to cope with potential suits from dealers over vehicles’ defects. From the start, most automakers included broad clauses requiring dealers to maintain repair facilities, but they soon added two clauses restricting their liability. One concerned the limited nature of their warranties. Ford led the way: Its 1904 agreement reprinted the industry trade association’s standard warranty (adopted in 1902). The company would replace defective parts only for the first sixty days after the car buyer received the vehicle. Any decision as to whether a part was defective was subject to the manufacturer’s final approval. Should any car be altered without the manufacturer’s approval, then it was no longer backed by the sixty-day warranty.[95] Although not as detailed as Ford’s warranty, many firms, such as Jeffery, Reo, Cadillac, and Oakland, added clauses after 1905. Further, contracts denied the manufacturer’s liability for defects in suppliers’ parts. Cadillac stated that its 1908 warranty did not cover “parts not made by this Manufacturer. The dealer must make all claims for such defective parts to their respective makers.”[96]

Dealer agreements were a mixed proposition for dealers. On the one hand, manufacturers added provisions, such as exclusive territories, to foster long-term ties with individual dealers, a practice that Stewart Macaulay and Ian Macneil have called “relational contracting.”[97] On the other, as long as manufacturers could deny the relational dimension of their dealer contracts in court—by claiming a sales contract as a close approximation to a “discrete transaction”—they exercised power by shifting liability from themselves to their dealers. Kessler thus wrote that the sales contract let manufacturers achieve “considerable control over the process of distribution” but “without exposure to the burdens and responsibilities of an agency relationship.”[98]

Ford’s general attorney, Leslie B. Robertson, understood the value of the sales contract. In 1913, he declared to one plaintiff’s attorney that “Mr. Ruggles [a Ford dealer] is in no sense an ‘Agent’ of this company, nor has he any authority whatever to bind it or to receive or accept service of any papers for it.” Robertson added that the branch office would furnish a copy of the dealer’s “agreement” so as to make clear “there is no basis upon which this suit can be maintained as against this company.”[99] That year he also faced a damage suit from South Carolina. Robertson wanted the dealer to sign an affidavit stating that he was not Ford’s agent, but the dealer asked Ford to “hold him harmless from all suits brought against him for machines sold under the terms of the contract.” The Ford legal department refused to do this. Putting the firm’s risk in broad terms, officials declared that it would “render us liable for unauthorized statements or agreements made by him to customers.” Eventually Ford prevailed on the dealer.[100]

The automaker also tried to shift liability to its wheel suppliers. In 1914 Robertson noted that he had pending “several defective wheel cases.” In a letter to Ford’s Arkansas attorney, he declared: “We do not manufacture wheels, nor have we ever done so, but purchase them from reliable concerns and after doing so make every inspection possible.”[101] Unlike consumers who lacked a contractual relation with manufacturers, manufacturers’ contracts with their wheel suppliers presumably offered them some leverage. Ford’s attorney, for instance, leaned on wheel suppliers to cover half the cost of litigation.[102] The contract also applied to defects. In the case of Olds Motor Works v. Shaffer (Ky. Ct. App. 1911), Olds lost the suit, according to Robertson, because the defect was so “apparent … that the Olds Company must have had knowledge of it.”[103] But defects were also latent. Robertson wrote in regard to the Livesay lawsuit in 1914: “There is no possible means by which we could find a hidden defect, nor is there any known method of making a test which would disclose same.”[104]

If Robertson expected the courts to hold wheel suppliers liable under these conditions, he also understood the implications of MacPherson v. Buick. In 1914, after the Appellate Division had ruled for MacPherson, Ford’s local attorney urged officials to “note carefully” what Justice Kellogg had said regarding the “inspection and testing of parts … purchased from maker of good reputation etc. If your test is not sufficient to detect latent defects, under the holding in the last case mentioned, you would be liable to a third person for personal injuries arising from such defect covered by paint or anything else.”[105] That year Robertson told his Oklahoma attorney, “If the manufacturer can be held liable for hidden defects in goods which are not made by him, it would result in a great many damage suits and put most manufacturers out of business.”[106]

Through the franchise sales contract, automobile manufacturers severed contractual relations between themselves and car buyers; via provisions to avoid liability for defective parts or delays, they limited their obligations to dealers. Following the path charted by Ronald Coase, economic and business scholars have investigated the structure of the modern firm, wanting to know why some activities were integrated within a firm and others transacted through the market. Their answers vary, but have almost always been couched in economic terms—transaction costs, path dependency, sunk costs, or specialized (tacit) knowledge.[107] Yet these scholars have not followed through on Coase’s study of social costs to ask how costs incurred by the buyers of a firm’s product, consumers and dealers, might have affected the firm’s structure.[108] The privity doctrine offers an explanation. Among important influences in their decision to rely on a franchise method of distribution, managers sought to shift liability for defects in a new market where the technology was imperfect and relationships uncertain.

III. MacPherson v. Buick: Demonstrating Liability in Court

Unlike most injured car owners, Donald MacPherson secured the proof necessary to demonstrate that his car was defective, and thus he made possible Cardozo’s attack on the privity requirement. Efforts to hold corporations responsible for industrial and consumer injuries increased during the Progressive era. In the case of worker injuries, Arthur McEvoy found a shift in “common sense” ways of understanding the causes of accidents. Crystal Eastman, as secretary of New York’s Employer’s Liability Commission, focused her study on the structural character of the work place, such as the type of machines used or the speed of their operation. Viewed this way, accidents followed “from the employers’ active control over the workplace,” not workers’ errors. Eastman’s social science approach thus accounted for industrial injuries “in a new way that made new laws … both conceivable and politically feasible.”[109] Similarly, as Barbara Welke recounted, railroad and streetcar safety regulation followed from public investigations of private corporations.[110]MacPherson differed from these cases since there was no comparable public investigation of GM’s production methods. GM controlled access to its research and never admitted its relationship with Arthur D. Little in court. Jonathan Lurie argued that MacPherson’s lawyer, Edgar Brackett, was the quiet force behind Cardozo’s ruling.[111] Brackett was lucky because the defect concerned wooden wheels, a component based on older ways of understanding the technology that existed outside of GM’s control. That Brackett established the defect did not imply, however, that Cardozo’s ruling suddenly prompted automakers to test products where they had not done so prior to 1916. Viewed from the perspective of relational contracting, Cardozo singled out a manufacturer’s responsibility to consumers as he limited its ability to rely on suppliers and dealers for a product’s quality.

Donald MacPherson held that his accident and injuries had been prompted by the car’s defect: a wheel made of rotten wood. As he drove along a road at a moderate speed, the wheel’s spokes collapsed, causing the car to roll over and damage his arm, wrist, and eyes.[112] The case raised seemingly simple questions: Was the wheel rotten? Was Buick responsible?

Buick relied on two defenses.[113] First, of course, was that it was not liable since it had sold the car to the Close Brothers. Cardozo put this excuse aside, finding that though the dealer was the immediate buyer to whom the manufacturer was responsible through the sales contract, the dealer was the one party likely not to use the car. Second, Buick tried to shift the firm’s liability back to the wheel maker, claiming it had purchased its wheels from a reputable supplier. But the defendant’s own witnesses made this claim hard to stick. G. W. Durham, an engineer who had worked with many automakers since 1900, testified that the managers of car manufacturers “finally determine as to the kind of wheels they will put under the body.”[114] Charles Johnson, a wheel maker for thirty-nine years, declared: “After the wheels left the wheel maker and went to the automobile maker, the manufacturer could put on them a car of such weight as he pleases.”[115] In other words, the car maker determined the balance between the car body’s weight and the strength of its wheels. The court concluded that it was “too remote” to hold the maker of “component parts” negligent when Buick had a duty to inspect.[116] Having failed to shift liability forward to the dealer or backward to the supplier, Buick’s internal practices were open for legal review: How had it tested or inspected its wheels?

As part of its defense of reliance on a reputable supplier, Buick’s witnesses stated that they did not inspect wheels, implying that it was not possible to inspect them, and that in the early 1900s no automaker inspected its wheels. G. W. Durham testified: “Prior to the year 1909 there was nothing done, that I know of, by automobile manufacturers before they put the wheels under the body to determine whether the spokes of the wheels were rotten or not.” Durham cited Packard, Pierce, Thomas, Ford, and Cadillac as firms where he “[did] not know of any” testing.[117] E. D. Cook, who worked with a wheel manufacturer, said that his firm primed the wheels before delivery. He knew of no means for testing a wheel’s “strength or any mechanical test to find out the strength of the spokes, or the quality of the hickory.”[118]

The plaintiff, by contrast, was fortunate in bringing suit about a wooden wheel. He did not rely on modern standards to test the wheel. Instead, three local carriage makers testified on his behalf, each of whom claimed twenty or more years of experience in judging hickory, the hard wood used to make the wheels’ spokes.[119] George Palmer examined MacPherson’s wheel soon after the accident and found “the wood was brittle, coarse grained, such as you find in old trees.” He stated, “sound hickory, when it breaks, it brooms up, slivers up…. The fact that they were brittle and of poor quality indicates very little strength, not half the strength of sound wood.”[120] Adelbert Payne judged hickory by its heft and its grain, finding that “different angles of the spoke expose different portions of the grain.”[121] James P. Tittemore testified that whereas sound wood splinters or brooms, dead or dozy wood lacks this elasticity, and side strains cause it to break easily.[122] Their remarks were familiar ones in the industry. In a case about a defective wheel, Ford’s brief recalled witnesses who described the wood as “dozy and brashy, meaning timber that would break square off and not splinter, and further showing that wood in that condition would break much more easily and upon less strain than perfectly sound timber.”[123]

Buick objected to the carriage makers’ testimony, arguing that such tests were not feasible because suppliers painted the wheels. Yet, to assess hickory’s grain, one needed to see it, as Payne explained: “The test with the paint on is to scrape off sufficient paint to see the grain and fiber of the wood. It certainly is a perfectly feasible and easy test to determine the character of a spoke that way, and sure.” Alternatively, the wood could have been inspected before it was put into the wheels. The plaintiff’s witness, Otto Kleinfelder, had worked nine years as a tester and stated that the Thomas Motor Car Company had wheels shipped “in their natural wood.” He also described a hydraulic pressure test where pressure was put on the hub to ascertain its strength. Buick’s witness, E. D. Cook, stated he did not know of Thomas’s hydraulic pressure test and thought it unusual, given common practices in the industry.[124] Buick, in a sense, pleaded for sympathy by saying that it was no different from other firms engaged in a process of innovation and in 1909 lacked methods to test products.[125] The plaintiff’s carriage makers, by contrast, had provided a standard to judge the defective product, one that evaded GM’s control and one that members of MacPherson’s community could easily place within their daily lives.

It would be a misunderstanding to picture Cardozo’s ruling as a grand turning point in product testing.[126] Prior to MacPherson, automakers tested vehicles as part of their general effort to develop a viable product. By 1910, the SAE and the Forest Service had initiated tests of materials and products, and by 1911, GM had hired Arthur D. Little to evaluate inputs. Second, as several court cases weakened the privity requirement, lawyers may have tried to bolster the defense of manufacturers with evidence of inspections. The Ford Motor Company, for example, inspected wheels purchased from suppliers and relied on its inspections for its defense when J. M. Livesay sued the firm over a defective wheel. Although GM had conducted no tests in 1909, Ford had its tests underway by 1913, before Cardozo’s ruling and the 1914 Appellate Division’s ruling. Ford’s brief contended that its inspections “were the same universally made by reputable manufacturers of automobiles.”[127] Its attorney thus distinguished his firm from Buick: “You will note in the Buick case no inspection or examination of the wheels was made and this seemed to be the strong point of the court in sustaining the verdict, while, in our case, we show at least nine separate and distinct inspections.”[128]

The issue at stake was not simply a question of testing; MacPherson drew attention to the “bright” and “dark” sides of relational contracting.[129] In the market’s early years, automakers had counted on ties with dealers as well as relations with suppliers who made many components that firms like GM and Ford assembled.[130] Yet, both GM and Ford exercised power through their relational contracts as they tried to shift liability for defective vehicles backward to their suppliers or forward to their dealers. Although some firms, like GM, benefited from their extended ties to research consultants, such as Arthur D. Little, in court the supplier or automaker needed to demonstrate tests such as Ford’s wheel inspections conducted in 1913 (and perhaps earlier). Yet, Ford’s managers still complained that for “latent or hidden defects” the firm would have been required to “apply extraordinary tests, which would mean to take the wheel apart, remove the spokes, scrape off the paint and apply tests to the wood, this being practically a complete destruction and rebuilding of the wheel.” Like Buick, Ford counted on its relations with dealers in order to call into play the privity requirement. In its case against Livesay, aside from its contention to have demonstrated “ordinary care” thanks to its many inspections, Ford’s focus on the privity requirement won the Oklahoma judge’s support.[131]

As MacPherson had located a means to establish the product’s defect outside of GM’s control, Cardozo’s decision built on earlier rulings and fit with later cases. The U.S. Appeals Court, for instance, reiterated Cardozo’s message in Johnson v. Cadillac Motor Car Co. (2d Cir. 1919), holding manufacturers responsible for inspecting their products.[132] Neither Cardozo nor the second circuit court, however, asked what research methods a firm needed to use to determine defects. It was a difficult question, but in the next several years, different parties established standards to monitor product quality and shape corporate research. Manufacturers still counted on their relational ties with suppliers and dealers to produce and sell vehicles, yet they also faced regulatory scrutiny, in part but not solely due to MacPherson.

IV. Consumers’ Risks in a Mass Market

Between 1916 and 1926, automobile sales nearly tripled from 1.6 to 4.3 million vehicles. In 1910, just one percent of households owned an automobile, but the figure jumped to twenty-six percent in 1920 and peaked at sixty percent in 1930 before slipping back to fifty-five percent from the mid-1930s through the 1940s.[133] As Ford, GM, and Chrysler assumed command of the market during the 1920s, they sought to win consumers’ loyalty or repeat purchases and, as part of this effort, took steps to fashion more reliable, durable, and by extension safer vehicles.[134]MacPherson intensified these market pressures: Manufacturers reduced their chances of being sued and improved their chances of defending themselves successfully by making good on Cardozo’s demand that they undertake careful inspections and tests of their products. This summary, however, belies a more complex account of efforts to root out defects. In a mass market, managers juggled many business concerns and, as they struck deals between safety and other goals, some old defects persisted and new ones emerged. Even with the modern research lab, corporations faced choices about a vehicle’s safe design. Cardozo’s ruling encouraged firms to test and inspect components, but offered no further specifications. Instead, a variety of public and private oversight agencies assumed the job of monitoring defects.[135] The SAE, the USDA’s FPL, insurance underwriters, the Massachusetts motor vehicle administration, the Eastern Conference of Motor Vehicle Administrations, and state legislatures targeted components critical to a car’s safety—bumpers, carburetors, headlights, electrical equipment, and glass windshields. This web of private and public oversight gave specific content to MacPherson, shaping managers’ choices about corporate research in their pursuit of consumers’ loyalty.

As GM, Ford, and Chrysler (organized in 1925) grabbed the largest shares of the auto market during the 1920s, they joined the ranks of a handful of corporations funding large research laboratories.[136] Prior to the 1920s, some companies (like GM) had contracted with consulting firms (like Arthur D. Little, Inc.), but most labs were small and tested materials and parts. The National Research Council (NRC) found that just fifteen firms employed at least fifty researchers in 1921. Firms in competitive industries like textiles and lumber could not afford large labs. By contrast, in markets dominated by a small number of large corporations, such as chemicals and electrical machinery, management invested in research.[137] Autos fit this pattern. GM, for instance, had earned $8.6 million in profits on $49 million in sales in 1910, but by 1927 the company pocketed $262 million in profits after taxes on $1.3 billion in sales. That year GM reported a research staff of 260 personnel for the NRC’s survey. Although Ford volunteered no data, Chrysler listed 143 researchers. This was at a time when 44 out of 926 companies staffed labs with fifty or more researchers. In 1938, 120 firms supported labs of this size.[138]

GM’s commitment to research provided the institutional foundation for safer products. In 1919 GM hired as its director of research Charles Kettering, an inventor best known for the electric self-starter who brought credibility and imagination to his job.[139] In July 1921 Pierre S. du Pont, as GM’s president, singled out the importance of a vehicle’s reliability by instructing Kettering to assess how a new car would operate after 20,000 to 30,000 miles on the road “with a definite record of troubles.” That fall Kettering had tests underway.[140] In the next three years, GM established its Proving Grounds, which Stuart W. Leslie noted “substituted scientifically designed banks, gradients, and measuring instruments for the drama of open road trials.” Thus, what Cox of Arthur D. Little requested, Kettering achieved: the ability to systematically assess materials, design components, test their properties (strength, durability, flexibility), and review devices under controlled driving conditions. GM backed Kettering’s lab with a generous annual budget of $375,000 in the early 1920s and roughly $2 million at the onset of World War II.[141]

GM’s lab conducted a wide variety of research projects, some of which improved safety as they made cars more reliable. Crankcase ventilation offers one illustration. With crankcase dilution, as Leslie observed, consumers found that “their timing chains and other engine parts were broken or badly pitted.” Engineers traced the problem to water, which caused particularly sulfurous fuels “under the right conditions of dilution” to “produce a sulfuric acid that ate tappets, piston rings, and other engine parts.” Ventilators were introduced on the Cadillac in 1925 and soon thereafter on other makes.[142] This new feature was one of many changes. A crankshaft balancer, for instance, reduced vibration and lowered repair costs. Longer-lasting fan belts also lowered repair bills.[143] Such advances coincided with a drop in accidents measured on a per car basis. In 1913 (the first year in the data series), the number of deaths caused by auto accidents averaged 30.7 per 10,000 cars; by 1926 the figure had dropped to 10.6 per 10,000 cars.[144]

Corporate initiatives alone could not take credit for this decline in accidents, however. MacPherson, as it built on related cases, offered an added inducement to inspect and test components. Both the auto and insurance press took note of Cardozo’s ruling.[145] In his survey of nearly all products liability cases involving automobiles at the appellate level from the early 1900s through the 1950s, Cornelius Gillam reported that in response to MacPherson a firm’s “most obvious defense” was evidence “that reasonable care was in fact used in the manufacturing process.”[146] That standard, he continued, stiffened over time, but the message that corporations could defend themselves by demonstrating proper care in inspections was not lost on the leading manufacturers. By 1916, as noted, Ford’s legal department had already included inspections and tests in its defense.[147] Eleven years later, in 1927, when Ford faced another damage suit for a defective wheel, the wheel supplier provided thorough tests for the year the wheel was made and described a testing lab, a chemical lab, and a lab for heat treatment tests among other tests.[148] Overall, Gillam reported that by the late 1920s automakers had created “their own inspection systems in wheel manufacturing plants,” and as time passed, they “continued to refine their inspection systems.”[149]

Behind this picture of corporate research and inspection a more complex process shaped efforts to eradicate technological defects. Managers sought consumers’ loyalty (repeat purchases) in a mass market, but the goal of making cars more reliable, and by extension, safer, ran up against other goals, such as reducing production costs or marketing new features. In surveys from the early 1930s, consumers expressed their concern for safety, as one study noted: “The necessity for greater safety is frequently mentioned.” In 1933, another study found that ninety percent of car buyers wanted safety glass. Consumers also focused on components easily taken for granted. They asked that tire valves be improved to stop air from leaking and that bumpers be positioned at “uniform height from [the] ground.”[150] Problems in the production of vehicles also resulted in defects. Citing clutches that “chattered,” GM’s Henry Crane declared in 1934: “The real trouble is that the clutches are designed with practically no factor of safety, solely to save a small amount of cost.”[151] In 1941, GM managers cited as “evidence” of Buick’s “lowered quality” its many engine problems plus “the number of broken steering gears experienced on the 1941 models as a result of steering gear misalignment.” Top officials debated whether the Central Office’s inspectors should “shut down the assembly lines” if a “desired standard” was not reached. They also proposed setting up “another million dollars and give Buick dealers a liberal allowance to correct all difficulties.”[152]

New risks of injury followed from marketing innovations, notably the major innovation of the early 1920s—enclosed cars.[153] By keeping out bad weather, enclosed cars became enormously popular, yet the new design created three major problems. The first concerned the quality of lumber. Through most of the 1920s, body builders stretched steel skins over wood frames. Enclosed vehicles demanded much stronger wood than open cars so as to prevent their beams from warping or breaking. Body builders, for instance, prized ash for beams, but complained that “brashy ash” (a light, weak wood) was difficult to distinguish from sound ash, and “a certain amount often gets by and into the finished body, where it occasionally fails in service under conditions which would not ordinarily cause breakage.”[154] The second problem was carbon monoxide poisoning. Because glass windows sealed a vehicle, Travelers Insurance found that “over 60 per cent. of all cars of all kinds … contain measurable concentrations of carbon monoxide, and that 7 per cent. of them contain quantities that may cause collapse.” The report blamed poor maintenance and defects. Where leaks in manifolds and mufflers or “a defective car heater” let carbon monoxide escape, gaps in floor boards directed the gas into the passenger compartment.[155] Third, glass often broke and cut passengers. Some firms installed safety glass (two pieces of glass bonded by a plastic sheet) by the late 1920s, but GM’s Alfred Sloan declared in 1932 that he could not justify the expense to shareholders.[156] Henry Ford, by contrast, adopted safety glass for the Model A. His decision was not a sales gimmick: Ford made this choice after broken glass injured an employee.[157] Yet, Ford faced a different tradeoff. A 1932 memo recalled that “improper lamination” had caused “separation of the glass, strain cracks, discoloration, bubbles, etc.” From 1928 to the end of May 1930, twenty percent of windshields were rejected as defective and nearly four million were replaced.[158] Sloan framed his response to safety glass relative to production costs, while Ford to the problems of innovation.

Automobiles received insurance underwriters’ considerable attention and offered one example of the many products subject to the inspections of the Underwriters’ Laboratories (UL). In his 2003 dissertation, Scott Knowles argued that the UL’s authority turned on “its ability to anticipate risks presented by new technologies, manage risks presented by existing technologies, and publicize fire risk not as a mysterious force of nature but as a calculable and tractable scientific problem.” The UL, Knowles reported, tested products and tracked them through their labels that indicated a product’s safety.[159] In 1918 the UL established an automobile council and underwriters worked with the SAE to establish categories for evaluating fire and collision hazards. Some eighty-five items, for instance, figured in the evaluation of fire hazards. Vehicles could receive up to a possible 8,000 points for their overall score and were graded in ten categories.[160] The new system surprised Ford in 1921 when its dealers protested that insurance rates on Model T’s had skyrocketed fifty percent. One dealer quoted from an insurance company’s letter: “‘Fire rates for Ford Automobiles have been materially advanced on 1921 models because of serious mechanical and structural defects, mainly in wiring systems.'” The letter continued: “‘These defects would be conducive to fire from short circuits, and in fact, underwriters have already experienced frequent and serious losses because of this.'” Whereas competitors’ makes had not experienced such rate increases, the dealer wrote in a huff that a Ford owner would need to pay $120 in insurance and that the high rates hurt his sales.[161]

Throughout the 1920s, the SAE established tougher standards. In the case of bumpers, for instance, in 1924 the SAE requested that bumpers be set at a uniform height, such that when two cars met, one bumper would hit the other car’s bumper, not its body.[162] Among government agencies, the USDA’s FPL graded maple, elm, hickory, red gum, and many other types of trees according to their “strength as a beam stiffness or post, shock resisting ability, and hardness.”[163] The SAE also worked with the Bureau of Standards to design equipment and test components. In the case of brakes, for instance, the Bureau developed the decelerometer to accurately measure the ability of brakes to stop vehicles.[164] The SAE published new codes in its journal. Further, the UL conveyed stricter standards through its inspections of many components.[165] It also published lists of its approved automotive devices, naming the maker and specific components. Bumpers, carburetors, fire extinguishers, lamps, and windshield wipers were all listed by the UL.[166]

This oversight process worked to corporations’ benefit in many cases. Compared to Ford’s insurance problems due to fire risks in 1921, by 1923 GM’s in-house study of the Model T reported: “The insulation has been made heavier and all wires are enclosed in heavy looms where it is required that they pass points where chaffing is likely to take place. The installation of this new style wiring has also reduced [Ford’s] insurance rates.”[167] In 1924, insurance rates again influenced Ford executives. Contemplating the sale of Fordson tractors for use in commercial buildings, the experimental engineer, W. T. Fishleigh, saw the need for the UL’s approval since “the question of general fire insurance risk always comes up when use of such tractor equipment is proposed.” The UL’s George Becker asked Fishleigh to alter the tractor’s design, including moving the layout of the fuel line such that gasoline leaks would “not impinge on electrical equipment or exhaust lines.” Once modified, the Fordson was approved for use in “warehouses, factories and the like, without [an] increase in standard fire insurance rates.”[168]

In taking a constructive approach, Ford was rewarded with an insurance rating that expanded a valuable market, but the influence of these oversight entities was limited. The FPL freely disseminated information throughout all wood-using industries, but claimed no authority over private manufacturers.[169] It could not compel automakers to buy wood of a given grade. Underwriters’ standards were important, but not foolproof. The UL rated components and, as Knowles explained, established an arbitration process for disgruntled clients. Still, appliances could pass UL tests yet earn poor marks and present serious risks. In the mid-1920s, flimsy headlamps and defective brakes fit this category. The SAE’s codes were not compulsory, and like the UL and FPL, it could not require improvements.[170]

For critical components, state regulators directly shaped corporate research. Lighting systems offered one example. In 1921, motor vehicle commissioners from ten eastern states formed a conference to assess lighting devices. Working with the Electrical Testing Laboratories (ETL), the conference selected twenty-two lighting systems that followed the 1922 standards established by the Illuminating Engineering Society (IES).[171] In response, by 1921 Ford had the ETL test its headlamps subject to the requirements for states that had endorsed the IES’s standards.[172] GM’s report of 1923 complimented Ford’s “‘H’ headlight lens” for providing “[e]xcellent road lights without glare.”[173] But events took a turn for the worse in 1925. The registrar for Motor Vehicles in Massachusetts, Frank Goodwin, warned Ford and other firms that the “construction [of their lights] is so flimsy as to render the headlamp unsuitable for practical use.” He added: “we are going to insist upon rigid enforcement of the lighting laws” and, if need be, the state would “compel owners of automobiles to replace defective headlamp equipment if it is not … in compliance with our law.” That fall Goodwin’s engineer in charge of electrical equipment issued a bulletin with suggestions for proper construction.[174]

In 1925, regulators also rejected GM’s headlamps. When the ETL tested the headlamps, the engineer J. H. Hunt reported, the deflected beam proved too difficult for GM’s officials to adjust and, as Hunt admitted, the adjustment was “too technical a matter for the average owner to carry out successfully.” Having failed to obtain New Jersey’s regulatory approval, GM failed to obtain the Eastern Conference’s approval. The company, Hunt suggested, could threaten “legal action,” but doing so meant “having to contend with persistent opposition in the future.” He instead proposed that GM replicate the ETL testing devices so that officials could be confident of the results when ETL tested its equipment. Hunt also wanted the company to improve its lighting devices, since better headlamps would be “very easily demonstrated” to consumers and thus “should be of real sales advantage.” In December, he further suggested that GM produce its own headlamps. Although he anticipated no cost savings from doing so, Hunt wanted the ability to control the quality of headlights as a step to assuring regulatory approval.[175]

Regulatory oversight also extended to brakes and safety glass. Massachusetts’s registrar, Frank Goodwin, reported to makers of trucks in 1925 “that 75% of the trucks in use had defective brakes and many of them could not be repaired so as to stop the vehicle within the required distance.” The decelerometer, the apparatus developed by the Bureau of Standards, had enabled regulators to identify defective brakes and enforce the state’s law. Telling managers that they could “avoid difficulty with their customers,” Goodwin advised them to make their brakes “sufficiently strong and durable to stand up in service.” He offered some “leeway” to allow firms time to redesign and resubmit their brakes, but warned that his twenty-two state inspectors were checking trucks and prepared to identify defective equipment.[176] Broken glass posed another serious hazard. In 1931, a sales manager for Pittsburgh, Plate & Glass sent the Ford purchasing manager a survey indicating that twenty-two percent of Ford owners or family members had been in accidents and that glass had injured nearly half of those involved in accidents.[177] Ford had faced the difficult problem of overcoming defects in manufacturing safety glass, whereas other manufacturers, notably GM, had failed to use safety glass. By 1934, as the historian Joel Eastman recounted in his study of automobile safety, states began passing laws requiring that safety glass be used in cars. Their campaign resulted in a compromise: safer glass for the front windshield, a cheaper, but somewhat inferior glass for side windows.[178]

In his review of product defects, the information economist Michael Spence pictured three options for state intervention: the state could undertake a campaign “to inform consumers”; it could impose liability on producers “above the levels that are voluntarily undertaken”; or it could regulate a product’s design.[179] My account puts the second two options in historical perspective for the case of automobiles. Large firms like GM voluntarily invested in R&D;, wanting to sell more reliable and safer vehicles as part of their goal of securing consumers’ loyalty in a mass market. But this overall goal obscured the conflicts that managers faced between a vehicle’s safe design and the drive to lower production costs or market new devices. MacPherson altered managers’ calculus as evidenced in their inspections and tests. Managers also absorbed regulators’ lessons to varying degrees. Responding to underwriters, Ford improved the Model T’s wiring system and lowered its insurance rates. Responding to state regulators, GM’s Hunt thought it prudent not to challenge authorities but to improve and market GM’s headlamps. Yet, safety glass also reiterated the message that GM and Ford struck deals between safety and other goals; this distinctly dangerous component further indicated the state’s role for imposing minimum standards. Finally, regulatory oversight remained imperfect. As consumers sued automakers, Gillam found that the outcome of their cases depended in part on the technology in question. Some components, like wheels or glass, were much easier to establish as defective than other components, such as transmissions.[180] Because defects persisted and liability rules stiffened, automakers took another step in seeking products liability insurance.

V. Products Liability Insurance

At the time of Cardozo’s ruling, the market for products liability insurance was new and small. The insurance had the potential to provide firms with one means of making losses due to liability for defects manageable and predictable, and automakers were among the first purchasers of this coverage.[181] Their reliance on products liability insurance depended on other firms also seeking coverage. At first, however, insurance underwriters faced the problem of adverse selection that resulted in large losses relative to their premiums and threatened to stall the market’s growth. This situation changed as stricter legal standards plus an increased public awareness and willingness to sue companies broadened the mix of firms seeking coverage.

By 1916, automakers had begun acknowledging Cardozo’s ruling by seeking insurance for product defects. Again, Ford’s experience offers a valuable example, and, as had been the case with product testing, its desire for insurance predated MacPherson. In a lawsuit concerning a defective wheel, Ford’s attorney wrote in 1916 that the Royal Indemnity Company now “carr[ies] our insurance risks against suits for alleged defective parts. Our insurance covers all cars manufactured and sold for the past three years and all claims of this nature should be promptly reported to the Royal Indemnity Company.”[182] Apparently, Ford’s legal department had misunderstood the nature of its coverage initially. In an earlier case, dating to 1914, Robertson had expressed the view that its public liability insurance covered defective repairs. But the underwriter quickly corrected him, emphasizing that it covered injuries to nonemployees on Ford’s premises, not injuries caused by defective repairs or products.[183] But by 1916, Ford had obtained public and products liability insurance, and in succeeding years its lawyers worked with insurance companies in handling suits.[184] In a case involving a broken wheel for a Model T purchased in 1915, Robertson requested and the Royal Indemnity Company agreed to settle this case (along with two others) and have Ford reimburse the underwriter. The claim examiner wrote: “We believe that a settlement of $70.00 in these three cases is cheaper than defending the suits.”[185] Again, in a case begun in 1924 over a defective wheel, Ford worked with its underwriter. The plaintiff demanded $50,000; Ford now had coverage from the General Accident Insurance Company for up to $10,000. The attorneys involved in Ford’s litigation proposed “that an adjustment out of court might be the wisest and most expedient course to pursue.” They accomplished this goal in late 1927 for $2,500.[186]

Ford had been among the first U.S. companies to obtain this type of coverage. In his 1953 dissertation, John McTeer Briggs stated that the field “originated in England sometime between 1890 and 1900 as a result of a series of poisoning cases.” But the field remained small. Although Ford obtained insurance in the 1910s, underwriters first officially recorded premiums for products liability as a distinct subfield of insurance in the amount of $115,000 in 1924.[187] Premiums jumped to $1 million by 1933. Further, despite the Great Depression, premiums tripled to $3.6 million in 1937 and then rose to $6.5 million in 1944.[188] Lacking records for manufacturers or underwriters, I cannot offer a detailed chronology of which firms sought coverage. For instance, some firms became self-insurers; by 1937 Ford relied mostly on self-insurance.[189] Still, the trade press suggests how underwriters addressed the problem of adverse selection in order to foster the new market.

In 1934 James M. Cahill of Travelers Insurance surveyed the new field, complaining of large, indeed unacceptable loss ratios (the ratio of losses to premiums). Because of this, agents had not tried to sell this type of insurance; instead, a narrow range of companies, notably makers of “foods, cosmetics,” and goods for which retailers required insurance, requested coverage. The upshot was “a very adverse selection of business against the insurance companies.” Cahill nevertheless thought the demand for insurance would broaden thanks to an altered legal context. Recent court decisions had further chipped away at the privity requirement. Baxter v. Ford Motor Co. (Wash. 1932), he noted, did not involve negligence, but instead focused on a factual statement conveyed through mass advertising.[190] Legal oversight entailed other measures. “Violation of pure food and drug acts,” he wrote, was now taken as “sufficient to show negligence and permit a recovery, since these statutes are enacted for the public’s protection from the very harm suffered.” According to the Uniform Sales Act, now effective in thirty states, “[w]ritten guarantees, labels, etc., are express warranties.” Cahill also recognized that much uncertainty remained in assessing liability, as he hedged his bets in writing: “even today, injured persons face the possibility of a test case, since the more stringent rules imposing liability are not yet decided law in most jurisdictions.”[191]

Manufacturers and underwriters also reported a growing public awareness of the meaning of liability.192 In 1931, according to Procter & Gamble’s insurance manager, makers of food products were especially vulnerable to damage suits.[193] Public awareness widened beyond food products during the 1930s and was matched by new tensions between retailers and manufacturers. Kimball & Price, a New York retailer, reported that products liability claims averaged $133 per case in 1936.[194] As one response, large retailers and chain stores began demanding that any manufacturer that sold them products agree to a “hold harmless” clause in their contracts.[195] In addition, Woolworth’s, Sears, and Kresge’s began “insisting on product liability insurance” for manufacturers who wanted their business. The insurance broker A. P. Connor noted in 1937 that mass retailers demanded that manufacturers provide “them with certificates of products liability insurance, thereby extending the coverage to the store in the event of a claim being made against the latter.”[196]

Given these market conditions, underwriters had two options. First, they could conduct their own test of products. In 1928 Travelers set up a Chemical Engineering Laboratory in its Engineering and Inspection Division. One of its principal activities was to test products for defects and to study manufacturers’ production methods.[197] Second, underwriters could forgo the business. In 1937, one of the most senior agents, A. P. Connor, declared that he was “quite selective” and took roughly one in ten risks. Writing in 1953, Briggs noted that two underwriters prohibited certain products, including “cosmetics,” “fertilizers,” and “tires.”[198]

Given the problem of adverse selection, products liability insurance pointed to the circuitous route by which corporations registered a stricter legal climate. While initially a small range of firms demanded insurance, gradually more producers sought coverage. That mass retailers forced manufacturers to get insurance reveals a stricter climate. Underwriters, by testing products, identified and reduced risks. Yet, because agents like Connor claimed to turn away so much potential business, they seemed unable to control defects. Insurance underwriters, then, set parameters by which technological defects could be lowered to the point of being insured, and in this way, they facilitated the growth of many, but not all markets.


MacPherson’s injuries provide one example of the social costs inherent in the process of market innovation. Unable to perfect the technology but able to sell machines in a market context, firms like Buick initially marketed highly imperfect motor cars as they worked to better their products. Recognizing that social costs were inherent in the process of innovation highlights the importance of law in addressing the economic question of who absorbed the costs of innovation in the auto market. Liability rules had been critical to the choices managers faced between exercising power in the market place to saddle car buyers with many costs or to invest in resources to produce safer machines. Conversely, the focus on social costs has revealed the importance of economics in addressing the legal question of what impact MacPherson had on automakers. Cardozo’s ruling was enmeshed in a broad oversight network of private and public entities. Managers incorporated this regulatory oversight into their calculus in the particular context of a mass market where they sought consumers’ loyalty.

Although overlooked in the many accounts of the auto industry, liability for defects and liability rules had shaped the market in two important regards. First, manufacturers had counted on the privity requirement to distance themselves from consumers’ lawsuits. It is conventional among economic historians to equate corporate power with a firm’s size or market share.[199] Yet, despite being small firms in a young market, automakers exerted considerable power through the privity doctrine when they shifted many start-up costs to dealers and consumers. Ironically, as Ford and GM increased in size and garnered large slices of the market, they sought to retain repeat customers and devoted a portion of their larger profits to researching, testing, and inspecting vehicles. Second, in contrast to the conventional focus on the drive to reduce production costs or transaction costs, liability for defects influenced the structure of the modern firm in the auto market. Managers relied on franchised retailers rather than sales agents. That is, they structured their method of mass distribution so as to sever their contractual ties with consumers. Stewart Macaulay emphasized that manufacturers added formal provisions to contracts and initiated informal policies to foster on-going relations with their dealers. Yet, officials also counted on the sales contract to deny these on-going relations in court for the sake of avoiding liability for defects. Managers had good reason to worry, because they sold such crude motor cars, and it was in this context that they opted to structure their firms in terms of franchised retailers, or what Friedrich Kessler called “vertical integration by contract.”[200]

Cardozo’s ruling signaled a new standard of corporate responsibility, although what that standard meant in practice depended on the specific context of the auto market. Thus, whereas legal scholars picture MacPherson in relation to other products liability cases, I treat Cardozo’s ruling as one ingredient among market and nonmarket factors that brought about a change in managers’ approach to product quality. The coming of a mass market had instilled in corporate managers a new dedication to reliability and safety. With a thirty percent stake in the market, companies like GM and Ford needed to secure consumers’ repeat purchases for their continued profitability. Further, in response to Cardozo’s ruling (and other decisions), managers established inspection systems and improved testing procedures. Yet, MacPherson lacked the specificity to assure that safety received top priority when managers themselves faced tradeoffs between safety and other business objectives. During the 1920s, as the market matured, several private and public entities focused managers’ attention on quality. Underwriters, for example, through their rating system and laboratory, encouraged managers to take a constructive approach in the numerous details in a vehicle’s design. The SAE consulted with federal research organizations, such as the Bureau of Standards, as it established uniform codes. Moreover, state regulators mandated new standards and checked equipment. Managers included this regulatory oversight in their calculations of how to balance safety in relation to other goals, notably cost savings and marketing innovations. GM offered one example in the case of headlights, Ford another for its wiring system. MacPherson had focused managers’ attention on product quality, but the overall objective of securing consumers’ loyalty subsumed conflicting goals and managers persisted in making choices that resulted in defects.

The many efforts to design safer products did not, then, root out all flaws in the complex technology. In succeeding years, new suits, court rulings, and a finer regulatory mesh sought to catch more defects.[201] This stricter regulatory oversight created management’s demand for products liability insurance. Yet, insurance offered a partial solution. Underwriters put pressure on manufacturers to reduce defects, but did not—indeed could not—compel management to eliminate the tension between safety and other business goals. Put another way, products liability insurance freed managers from having to resolve the tradeoff between safety and the defects that were by-products of efforts to lower production costs or market new features. Therefore this insurance facilitated the growth of mass markets. Technological defects were brought under stricter oversight. But there was no perfect resolution in terms of the courts, public regulation, or private insurance.

This account of the auto market brings us back to the topic of innovation. Economic scholars have substantially revised Schumpeter’s concept of innovation as well as his focus on the entrepreneur. While unusually creative individuals remain vital to an economy’s development, the “entrepreneur” lives on not simply as a theoretical concept but also as an ideological construct. True to his conservative disposition, Schumpeter claimed that these risk-loving individuals deserved to be richly rewarded for boosting the economy’s level of productivity and society’s standard of living.[202] Popular culture continues to celebrate this view of entrepreneurs along with other risk-takers—the wildcatter, the derrick man, the futures trader. In praising the entrepreneur for assuming huge risks, the image has had another effect: It has obscured or effaced a different picture of the risks of innovation—risks that were widely distributed and absorbed by various members of society.

Sally H. Clarke is an associate professor of history at the University of Texas at Austin. She especially wishes to thank Christopher L. Tomlins, and to extend her many thanks as well to Naomi R. Lamoreaux, L. B. Boyce, Nelson Bowers, Jack Brown, Lizabeth Cohen, Mary Butler Davies, John Franz, Louis Galambos, Bruce Hunt, Peter Jelavich, Alice Kessler-Harris, Scott G. Knowles, James R. Sahlem, Steven W. Usselman, Barbara Welke, and the journal’s anonymous referees. Many librarians and archivists aided her research, including David White at the Scharchburg Archives of Kettering University, Cathy Latendresse and Linda Skolarus at The Henry Ford, Sarah Roberts at Michigan State University, Mark Patrick at the Detroit Public Library, Jo Ann Mattingly of the Supreme Court Library at Buffalo, and the librarians at the National Agricultural Library, the Library of Congress, the University of Texas at Austin, and the many state and federal archives. Much of the research for this article was undertaken while she was a fellow at the Radcliffe Institute for Advanced Study at Harvard University, and she wishes to thank the dean, Drew Gilpin Faust, and the director of fellows, Judith Vichniac. The Alfred P. Sloan Foundation and the University of Texas at Austin provided financial support that, at the margin, was critical to the completion of her research.



1. George H. Close, testimony, 36–37, quote 37, Donald C. MacPherson v. Buick Motor Company, Court of Appeals of the State of New York, March 14, 1916, MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050 (N.Y. 1916), Supreme Court Library at Buffalo, Buffalo, New York (hereafter Records and Briefs for MacPherson).

2. MacPherson’s accident is described in MacPherson v. Buick Motor Co., 138 N.Y.S. 224 (N.Y. 1912), 225; Complaint, 3–7, and Donald C. MacPherson, testimony, 15–20, quote 16, Records and Briefs for MacPherson. The accident is also described in David W. Peck, Decision at Law (New York: Dodd, Mead & Company, 1961), 41–42.

3. Donald C. MacPherson, testimony, 21–24, quote 21, Records and Briefs for MacPherson; Complaint, 5, Records and Briefs for MacPherson.

4. MacPherson (1912); Statement, 1–2, and Judgment of March 10, 1913, 12, Records and Briefs for MacPherson.

5. MacPherson v. Buick, 145 N.Y.S. 462 (N.Y. 1914), 465.

6. MacPherson v. Buick, 111 N.E. 1050 (N.Y. 1916), 1055.

7. MacPherson (1916). The legal literature concerning Cardozo is abundant. Edward H. Levi, An Introduction to Legal Reasoning (Chicago: University of Chicago Press, 1948), 7–19; Steven P. Croley and Jon D. Hanson, “Rescuing the Revolution: the Revived Case for Enterprise Liability,” Michigan Law Review 91 (1993): 683–797; John C. P. Goldberg and Benjamin C. Zipursky, “The Moral of MacPherson,” University of Pennsylvania Law Review 146 (August 1998): 1733–1847; Andrew L. Kaufman, Cardozo (Cambridge: Harvard University Press, 1998), 265–85; Jonathan Lurie, “Lawyers, Judges, and Legal Change, 1852–1916: New York as a Case Study,” Working Papers from the Regional Economic History Research Center, ed. Glenn Porter and William H. Mulligan, Jr., 3 (Wilmington, Del.: Eleutherian Mills-Hagley Foundation, 1980), 31–56; and William E. Nelson, The Legalist Reformation: Law, Politics, and Ideology in New York, 1920–1980 (Chapel Hill: University of North Carolina Press, 2001), 93–107, 187–88. Distinctly valuable for its extensive review of automobile cases is Cornelius W. Gillam, Products Liability in the Automobile Industry: A Study in Strict Liability and Social Control (Minneapolis: University of Minnesota Press, 1960). Valuable in linking Cardozo’s ruling to engineers’ ideas about managers and industrial accidents is John Fabian Witt, “Speedy Fred Taylor and the Ironies of Enterprise Liability,” Columbia Law Review 103 (2003): 1–49.

8. George H. Close, testimony, 36, Records and Briefs for MacPherson.

9. In their analysis of the privity requirement as it related to products liability, legal scholars typically have singled out the English case, Winterbottom v. Wright, 10 M & W. 109 Eng. Rep. 402 (Ex. 1842), in which a mail coach’s driver was injured when the coach broke and overturned. The driver sought compensation not from his employer, the Postmaster General, but instead from the company that had contracted with the postmaster for maintaining the coach. The court declared that the plaintiff could not recover since he was not in privity of contract with the coach’s manufacturer. In the United States, the courts soon began finding exceptions to the privity requirement. In 1858, for a New York case in which the defendant had mislabeled a bottle of belladonna (a poison) as a bottle of dandelion extract, a safe liquid, and sold the bottle to a druggist who in turn sold the bottle to another retailer who sold the bottle to the ultimate user, who was injured when she consumed the contents. The court, recognizing that the sale of the poison had been made to a druggist, reasoned that “The injury therefore was not likely to fall on him, or on his vendee who was also a dealer; but much more likely to be visited on a remote purchaser, as actually happened.” Finding that the defendant’s actions put the plaintiff’s “life in imminent danger,” the court took exception to Winterbottom and affirmed the judgment for the plaintiff. Thomas v. Winchester, 6 N.Y. 397 (1852), 409. The courts spent several decades enlarging exceptions to the privity requirement. See Devlin v. Smith, 89 N.Y. 470 (1882); Huset v. J. I. Case Threshing Mach. Co., 120 F. 865 (8th Cir. 1903); Kuelling v. Roderick, 75 N.E. 1098 (N.Y. 1905); and Statler v. Ray Manufacturing Company, 88 N.E. 1063 (N.Y. 1909). Among legal reviews of this topic, see Levi, An Introduction to Legal Reasoning, 7–19; Lurie, “Lawyers, Judges, and Legal Change, 1852–1916,” 31–56; Croley and Hanson, “Rescuing the Revolution,” 695–97; and Goldberg and Zipursky, “The Moral of MacPherson,” 1750–52.

10. MacPherson (1916), 1053. William L. Prosser, Handbook of the Law of Torts (St. Paul, Minn.: West Publishing Co., 1941), 677; Croley and Hanson, “Rescuing the Revolution,” 697–98. See also Goldberg and Zipursky’s critique of Prosser in “The Moral of MacPherson,” 1756–69.

11. R. H. Coase, “The Problem of Social Cost,” Journal of Law & Economics 3 (1960): 1–44. Writing about autos, Gillam included among social costs the death and injury of persons as a result of defective vehicles, the value of their labor that was lost, any medical care, plus the resources that went into the vehicles that were lost as a result of their defects. See Gillam, Products Liability in the Automobile Industry, 196–210, especially 196–97. Arthur F. McEvoy examined social costs in The Fisherman’s Problem: Ecology and Law in the California Fisheries, 1850–1980 (New York: Cambridge University Press, 1986). Barbara Welke found that advances in technology plus the increased speed of trains created new risks of injury for passengers and bystanders. Barbara Young Welke, Recasting American Liberty: Gender, Race, Law, and the Railroad Revolution, 1865–1920 (Cambridge: Cambridge University Press, 2001), 27–28. Scott Knowles focused on dangerous products and the risk of injury stemming especially from fire hazards. Knowles examined how private oversight entities, notably the Underwriters’ Laboratories, developed bodies of knowledge and the authority to influence corporate managers and shape their markets. Scott Gabriel Knowles, “Inventing Safety: Fire, Technology, and Trust in Modern America” (Ph.D. diss., Johns Hopkins University, 2003). Sarah S. Lochlann Jain examined the injuries borne by bystanders in the early auto market and the mature market of the 1950s and 1960s in “‘Dangerous Instrumentalities’: The Bystander as Subject in Automobility,” Cultural Anthropology 19 (2004): 61–94. Albert O. Hirschman studied the dynamics between buyers and sellers in mature markets when product quality deteriorates in Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Cambridge: Harvard University Press, 1970). John Fabian Witt, “Toward a New History of American Accident Law: Cooperative First-Party Insurance Movement,” Harvard Law Review 114 (2001): 690–841.

12. One of the first automobile historians, Ralph Epstein, wrote that through the year 1910 “the expense occasioned by the rapid breaking and wearing out of parts … often either equaled or exceeded the annual cost of operation.” Ralph C. Epstein, The Automobile Industry: Its Economic and Commercial Development (1928; reprint, New York: Arno Press, 1972), 85. On Ford’s first experimental auto, see Allan Nevins, Ford: The Times, the Man, the Company (New York: Charles Scribner’s Sons, 1954), 148–57. Pamela Walker Laird outlined automakers’ efforts to reassure motorists about their products’ quality in “‘The Car without a Single Weakness’: Early Automobile Advertising,” Technology & Culture 37 (October 1996): 796–812.

13. Flaws in the wood, hickory, were the defects at issue in MacPherson’s case. Manufacturers favored hickory, researchers reported, because “[t]he severe thrust, strain, twist, and compression which automobile wheels must sustain.” Still, defects in hickory and other species of trees took many forms, often called “brashness.” Arthur Koehler cited decay initiated by fungi as “a well-known cause of brashness” and explained: “brash wood breaks suddenly and completely across the grain with brittleness in fracture and with a comparatively small deflection.” “Shock resistance,” he concluded, was “the first mechanical property affected by the progressive disintegration of wood by fungi. Wood may show a reduction in this property even before the decay has advanced far enough to be readily recognized by inspection or before the type of fracture is affected by it.” Charles F. Hatch, “Manufacture and Utilization of Hickory, 1911,” U.S. Department of Agriculture, Forest Service—Circular 187 (1911): 1–16, quote 4; and Arthur Koehler, “Causes of Brashness in Wood,” U.S. Department of Agriculture Technical Bulletin 342 (1933), quotes 2, 36, 38.

14. As a cautionary note, I do not argue that liability for defects was the sole factor leading to the adoption of auto franchising. My complaint is that economic and business historians have overlooked liability, in general, and defects, in particular, as factors in managers’ decisions. Of most importance, liability for defective products (or other problems) is omitted in the major works in the field. See especially, Alfred D. Chandler, Jr., The Visible Hand: The Managerial Revolution in American Business (Cambridge: Harvard University Press, 1977); idem, Scale and Scope: The Dynamics of Industrial Capitalism (Cambridge: Harvard University Press, 1990); and Oliver E. Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting (New York: Free Press, 1985). Economists have written at length about franchising, but in my survey of the literature I did not find that they had examined the impact of liability on a firm’s structure. See for example, Antony W. Dnes, “The Economic Analysis of Franchise Contracts,” Journal of Institutional and Theoretical Economics 152 (1996): 297–324; Paul Rubin, “The Theory of the Firm and the Structure of the Franchise Contract,” Journal of Law and Economics 21 (1978): 223–33; and Patrick J. Kaufmann and Francine LaFontaine, “Costs of Control: The Source of Economic Rents for McDonald’s Franchisees,” Journal of Law and Economics 37 (1994): 417–53. The business historian Thomas S. Dicke identified liability as a factor in Ford’s reliance on the franchise contract, but did not explore the implications of liability for autos or firms in other industries. Dicke relied on the scholarship of Charles Mason Hewitt, Jr., who traced the evolution of dealer agreements, but his treatment of defects was flawed, which I discuss in my appendix. Thomas S. Dicke, Franchising in America: The Development of a Business Method, 1840–1980 (Chapel Hill: University of North Carolina Press, 1992), 66–67; and Charles Mason Hewitt, Jr., Automobile Franchise Agreements (Homewood, Ill: Richard D. Irwin, Inc., 1956), 37–38. The appendix is available at <//www.historycooperative.org/journals/lhr/23.1/clarke.html> (hereafter Appendix).

15. Stewart Macaulay, Law and the Balance of Power: The Automobile Manufacturers and Their Dealers (New York: Russell Sage Foundation, 1966); Ian Macneil, “Economic Analysis of Contractual Relations: Its Shortfalls and the Need for a ‘Rich Classificatory Apparatus,'” Northwestern University Law Review 75 (1981): 1018–61; idem, “Bureaucracy and Contracts of Adhesion,” Osgoode Hall Law Journal 22 (1985): 5–28; Friedrich Kessler, “Automobile Dealer Franchises: Vertical Integration by Contract,” Yale Law Journal 66 (1957): 1135–90; Gillian K. Hadfield, “Problematic Relations: Franchising and the Law of Incomplete Contracts,” Stanford Law Review 42 (1990): 927–92; and Robert W. Gordon, “Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law,” Wisconsin Law Review (1985): 565–79. Croley and Hanson singled out Kessler’s impact on “the first generation of product liability scholars and judges” in “Rescuing the Revolution,” 691, n. 29, quote 708–9. In his studies of corporate power, Hanson has combined the insights of legal realists and critical legal scholars. See Jon Hanson and David Yosifon, “The Situation: An Introduction to the Situational Character, Critical Realism, Power Economics, and Deep Capture,” University of Pennsylvania Law Review 152 (2003): 129–346. John Henry Schlegel explored realists’ empirical studies in American Legal Realism and Empirical Social Science (Chapel Hill: University of North Carolina Press, 1995).

16. Faulting recent legal writers, Croley and Hanson argue that consumers still face problems of imperfect information, and the other two complaints still apply to markets. Croley and Hanson, “Rescuing the Revolution,” 687–797, especially 767–97, quotes 690, 769. Further Jon D. Hanson and Douglas A. Kysar argue that firms manipulate consumers’ perceived risks of defects (often through mass media) in “Taking Behavioralism Seriously: Some Evidence of Market Manipulation,” Harvard Law Review 112 (1999): 1420–1572.

17. Lawrence M. Friedman and Jack Ladinsky, “Social Change and the Law in Industrial Accidents,” in American Law and the Constitutional Order: Historical Perspectives, ed. Lawrence M. Friedman and Harry N. Scheiber, enl. ed. (Cambridge: Harvard University Press, 1988), 269–82; Witt, “Toward a New History,” 690–841, and “Speedy Fred Taylor,” 1–49; Christopher L. Tomlins, “A Mysterious Power: Industrial Accidents and the Legal Construction of Employment Relations in Massachusetts, 1800–1850,” Law and History Review 6 (1988): 375–438; Lawrence M. Friedman and Thomas D. Russell, “More Civil Wrongs: Personal Injury Litigation, 1901–1910,” American Journal of Legal History 34 (1990): 295–314; Arthur F. McEvoy, “The Triangle Shirtwaist Factory Fire of 1911: Social Change, Industrial Accidents, and the Evolution of Common-Sense Causality,” Law and Social Inquiry 20 (1995): 621–51; and Welke, Recasting American Liberty.

18. Welke, Recasting American Liberty, 3–136.

19. McEvoy, “The Triangle Shirtwaist Factory Fire of 1911,” 621–51; and Tomlins, “A Mysterious Power,” 386–87.

20. Although not focused on innovation, a valuable account of automobile safety is Joel W. Eastman, Styling vs. Safety: The American Automobile Industry and the Development of Automotive Safety, 1900–1966 (New York: University Press of America, 1984). See also Jain, “‘Dangerous Instrumentality,'” 61–94.

21. Automobile historians typically have focused on the many ingredients by which manufacturers produced a complex machine on a mass scale. They have asked how producers improved devices or marketed goods, rather than who bore the costs of innovation. There are a few exceptions. The economist Michael Spence called attention to liability as an information problem. Cornelius W. Gillam in his review of appellate cases involving automobiles argued that consumers absorbed social costs in the form of injuries born out of defective vehicles. Among excellent scholarly accounts of the automobile market, see Alfred D. Chandler, Jr., Strategy and Structure: Chapters in the History of the American Industrial Enterprise (Cambridge: MIT Press, 1962); Donald Finlay Davis, Conspicuous Production: Automobiles and Elites in Detroit, 1899–1933 (Philadelphia: Temple University Press, 1988); David Hounshell, From the American System to Mass Production, 1800–1932: The Development of Manufacturing Technology in the United States (Baltimore: Johns Hopkins University Press, 1984); Arthur J. Kuhn, GM Passes Ford, 1918–1939: Designing the General Motors Performance-Control System (University Park: Pennsylvania State University Press, 1986); and Stuart W. Leslie, Boss Kettering: Wizard of General Motors (New York: Columbia University Press, 1983). On liability for defects, see Michael Spence, “Consumer Misperceptions, Product Failure and Producer Liability,” Review of Economic Studies 44 (1977): 561–72; Gillam, Products Liability in the Automobile Industry, 196–210; and Jain, “‘Dangerous Instrumentality.'”

22. Joseph Schumpeter, Capitalism, Socialism and Democracy (1942; reprint, New York: Harper & Brothers, 1976); and idem, “The Creative Response in Economic History,” Journal of Economic History 7 (1947): 149–59. Many scholars have revised parts of Schumpeter’s conceptual framework. An economic revision is found in Richard R. Nelson and Sidney G. Winter, An Evolutionary Theory of Economic Change (Cambridge: Harvard University Press, 1982). For a business historian’s approach, see Nancy F. Koehn, Brand New: How Entrepreneurs Earned Consumers’ Trust from Wedgwood to Dell (Boston: Harvard Business School Press, 2001). Treating a government agency as an entrepreneurial entity, see Arthur L. Norberg and Judy E. O’Neill, Transforming Computer Technology: Information Processing for the Pentagon, 1962–1986 (Baltimore: Johns Hopkins University Press, 1996).

23. My point is not meant as a criticism. Much of the best scholarship in the field examines innovation from this perspective. In addition to the sources cited in note 22, see, for instance, Louis Galambos with Jane Sewell, Networks of Innovation: Vaccine Development at Merck, Sharp & Dohme, and Mulford, 1895–1995 (New York: Cambridge University Press, 1995); David Hounshell and John K. Smith, Science and Corporate Strategy: Du Pont R&D;, 1902–1980 (New York: Cambridge University Press, 1988); W. Bernard Carlson, Innovation as a Social Process: Elihu Thomson and the Rise of General Electric (New York: Cambridge University Press, 1991); and Thomas J. Misa, A Nation of Steel: The Making of Modern America, 1865–1925 (Baltimore: Johns Hopkins University Press, 1995).

24. The literature about consumption is now vast. Business historians and historians of design, in particular, have written about managers’ efforts to respond to consumers, but have not examined consumers’ lawsuits. Two valuable studies are Koehn, Brand New, and Regina Lee Blaszczyk, Imagining Consumers: Design and Innovation from Wedgwood to Corning (Baltimore: Johns Hopkins Press University, 2000). Economic historians have studied accidents, but as a byproduct rather than as a central element of the economy’s development. See Peter Temin, Taking Your Medicine: Drug Regulation in the United States (Cambridge: Harvard University Press, 1980); and Price V. Fishback and Shawn Everett Kantor, A Prelude to the Welfare State: The Origins of Workers’ Compensation, National Bureau of Economic Research (Chicago: University of Chicago Press, 2000).

25. Ron Harris recently surveyed economic historians’ growing interest in legal history. Ron Harris, “The Encounters of Economic and Legal History,” Law and History Review 21 (2003): 297–346. See also Gordon, “Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law,” 565–79; Witt, “Toward a New History,” 692–713; Hanson and Yosifon, “The Situation,” 179–201; and Schlegel, American Legal Realism.

26. Tomlins, “A Mysterious Power,” 375–438; and McEvoy, “The Triangle Shirtwaist Factory Fire of 1911,” 621–51.

27. Richard P. Scharchburg, Carriages without Horses: J. Frank Duryea and the Birth of the American Automobile Industry (Warrendale, Pa: Society of Automotive Engineers, 1993); James J. Flink, The Automobile Age (Cambridge: MIT Press, 1988), 5–6, 23, 31–32; Editors of Automobile Quarterly, The American Car since 1775: The Most Complete Survey of the American Automobile Ever Published, 2nd ed. (New York: L. Scott Bailey, 1971), 138–39; and U.S. Federal Trade Commission, “Report on Motor Vehicle Industry,” 76th Cong., 1st sess., House Document No. 468 (Washington, D.C.: U.S. Government Printing Office, 1939), 22–23.

28. Flink, The Automobile Age, 31–35. See also The American Car since 1775, 138–39.

29. For industry developments, see Flink, The Automobile Age, 25, 37, 64, 212–13. On Kettering’s self-starter, see Leslie, Boss Kettering, 49–50. On the assembly line, see Hounshell, From the American System to Mass Production, 217–61. Among studies of the early automobile industry, one of the oldest and still most valuable accounts is Epstein, The Automobile Industry. See also John B. Rae, American Automobile Manufacturers: The First Forty Years (Philadelphia: Chilton Company, 1959); Nevins, Ford: The Times, the Man, the Company; Davis, Conspicuous Production; and Clay McShane, Down the Asphalt Path: The Automobile and the American City (New York: Columbia University Press, 1994).

30. Albert L. Clough, “The Question of Axles,” Horseless Age 8 (November 6, 1901): 672.

31. “Automobile Accidents,” Horseless Age 10 (August 20, 1902): 205.

32. “Lessons of the Road,” Horseless Age 9 (April 2, 1902): 449.

33. Nevins, Ford: the Times, the Man, the Company, 247–49, quote 247.

34. “Automobile Accidents,” Horseless Age 10 (August 13, 1902): 180.

35. N. B. Pope, “The Automobile as a Fire Hazard,” Weekly Underwriter 86 (February 3, 1912): 117–19. See also “Leaky Tanks,” Horseless Age 10 (July 9, 1902): 28.

36. “Automobile Insurance,” Automobile 11 (July 18, 1903): quote 62; and Dixie Hines, “The Insuring of Automobiles,” Automobile 11 (August 15, 1903): 160–61. On the history of automobile insurance, see Eugene F. Hord, “History and Organization of Automobile Insurance,” Speech Delivered before the Insurance Society of New York, November 11 and 18, 1919, part II, 15–29, Widener Library (stacks), Harvard University, Cambridge, Mass.

37. Weekly Underwriter 64 (May 18, 1901): 359.

38. “Difficulties of Automobile Insurance,” Horseless Age 9 (June 4, 1902): 667.

39. Hines, “The Insuring of Automobiles,” quote 160.

40. Weekly Underwriter 74 (January 6, 1906): 4. Prior to 1906, articles noted the demand for property coverage. See “Automobile Insurance Notes,” Automobile Topics 1 (October 27, 1900): 48; Hines, “The Insuring of Automobiles,” 160; “Automobile Insurance,” Automobile 11 (July 18, 1903): 62; “Automobile Insurance Rates,” Automobile 11 (September 19, 1903): 284; “Automobile Housing and Insurance,” Automobile 11 (November 28, 1903): 570; and “Blanket ‘Floater’ Insurance Rate,” Automobile 11 (November 28, 1903): 577.

41. “Automobile Liability,” Eastern Underwriter 9 (December 24, 1908): 14.

42. In the first decade of motoring, several auto journals appealed to wealthy buyers. See Automobile, The Automobile Magazine, Automobile Topics, and The American Automobile.

43. H. G. Buschman, General Merchandize, to “Friend Olds,” May 10, 1903, Folder 7, Box 6, Ransom E. Olds Papers, Michigan State University Archives and Historical Collections, East Lansing, Mich. (hereafter Olds Papers, MSU).

44. A few drivers were women. In 1903, Automobile counted fifty women among Boston’s 3,500 licensed motorists (who were not chauffeurs). “Four Thousand Boston Men and Women Licensed Motorists,” Automobile 11 (October10, 1903): 362. Virginia Scharff reported scattered data about women drivers. In Houston, Texas, she found that 5.5 percent of people owning cars in the city’s auto directory were women. Scharff noted that more women in all likelihood drove cars, but were not listed in directories. Virginia Scharff, Taking the Wheel: Women and the Coming of the Motor Age (New York: Free Press, 1991), 25–26. Welke detailed gendered distinctions in risk-taking among railroad passengers in Recasting American Liberty, 84–105.

45. “Record Breaking Tours,” Horseless Age 10 (July 23, 1902): 82.

46. Anonymous, “Diary Notes of a User, Part III,” Horseless Age 10 (October 1, 1902): 353.

47. W. E. Bristol, Cashier, Bank of Oakland, to E. W. Olds, March 3, 1903, Folder 3, Box 6; C. H. Farnum to Oldsmobile Co., March 6, 1903, Folder 3, Box 6; and E. W. Olds to B. C. Dinsmore, April 22, 1903, Folder 6, Box 6, all in Olds Papers, MSU.

48. Flink, The Automobile Age, 34.

49. Hounshell, From the American System to Mass Production, 234.

50. On the development of research laboratories among different industries, see David Mowery and Nathan Rosenberg, Technology and the Pursuit of Economic Growth (New York: Cambridge University Press, 1989), 79–92. On railroads’ laboratories, see Steven W. Usselman, Regulating Railroad Innovation: Business, Technology, and Politics in America, 1840–1920 (New York: Cambridge University Press, 2002), 199–211. See also studies cited in note 23.

51. See “Need of Testing Laboratories,” Horseless Age 17 (May 9, 1906): 657–58; and Harry E. Dey, “Experimental Departments,” Horseless Age 20 (October 2, 1907): 455–56.

52. Misa provided an excellent example of the SAE’s standardization work for steels used in automobiles. He noted that in 1917 the association changed its name to the Society of Automotive Engineers. Misa, A Nation of Steel, 213, 215–23, 229; and Rae, American Automobile Manufacturers, 79–80.

53. The Forest Service undertook many research projects for the wood-using industries prior to the creation of the FPL. Charles A. Nelson, “A History of the Forest Products Laboratory” (Ph.D. diss., University of Wisconsin, 1964), especially 24–27, 42–51, 57.

54. Mowery and Rosenberg, Technology and the Pursuit of Economic Growth, 84–90.

55. The research staff was described in the long letter from Vice-President, Arthur D. Little, Inc., to Thomas Neal, Esq., President, General Motors Company, June 1, 1911, Box 228, Arthur D. Little, Inc. Collection, Manuscripts Division, Library of Congress, Washington, D.C. (hereafter Arthur D. Little, LC). On standardization, see General Motors Company-Research Dept., Arthur D. Little, Inc.—Directors, “Progress Report of the Research Department for the Month of November, 1911,” 11 “GM Before 1923,” General Motors Company Files, National Automotive History Collection, Detroit Public Library, Detroit, Mich. (hereafter DPL); and Engineering Department, General Motors Company, Directors to Thomas Neal, President, GMC, “Progress Report of the Engineering Department for the Month of July, 1911,” 4–5, Box 228, Arthur D. Little, LC. For examples of projects undertaken by Arthur D. Little, Inc., see J. G. Callan, Engineer, Arthur D. Little, Inc., to General Motors Company, August 2, 1912, 711–40, Miscellaneous Technical Reports, Volume 21, Box 67; and General Motors Company—Engineering Dept., “Progress Report of the Engineering Department for the Month of September, 1911,” September 30, 1911, Box 228, both in Arthur D. Little, LC; General Motors Company, Research Department, Arthur D. Little, Inc., Directors, “Report of Research Department for July, August, and September,” No. 196, October 15, 1912, “GM Before 1923,” General Motors Company Files, DPL.

56. Engineering Department, General Motors Company, to Thomas Neal, Esq., President, General Motors Company, June 6, 1911, and “Test of Wheel 36″ × 5″ Oldsmobile Made by Imperial Wheel Company, March 15, 1911, At Flint Axle Works, Flint Michigan,” Miscellaneous Technical Reports, Arthur D. Little, Inc., Directors, Volume 18, 230–37, quote 230, 232, 237, Box 64, Arthur D. Little, LC.

57. Claude E. Cox to Gleason Murphy, December 27, 1911, 2, Box 6, Claude E. Cox Collection, Accession 15 (hereafter Cox), DPL.

58. “Increase in Number of Front Wheel Failures,” Horseless Age 23 (March 24, 1909): 398.

59. Claude E. Cox to Gleason Murphy, December 27, 1911, 2, Box 6, Cox, DPL.

60. Engineering Department, Directors, General Motors Company to Thomas Neal, President, General Motors Company, “Progress Report of the Engineering Department for the Month of June 1911,” July 1, 1911, 2, Box 228, Arthur D. Little, LC; General Motors Company-Research Dept., Arthur D. Little, Inc.—Directors, “Progress Report of the Research Department for the Month of November, 1911,” 11, 13, “GM Before 1923,” General Motors Company Files, DPL; Claude E. Cox to Gleason Murphy, December 27, 1911, 2, Box 6, Cox, DPL.

61. Claude E. Cox to Gleason Murphy, December 27, 1911, quote 1; Research Department, Arthur D. Little, Inc., [CEC] to Thomas Neal, President, General Motors Co., February 15, 1912; President, Arthur D. Little, Inc., to Tracy Lyon, Director of Production, March 28, 1912; Director of Research [Cox] to Arthur D. Little, Inc., August 9, 1912; and Director of Research to Arthur D. Little, Inc., attention Mr. John G. Callan, January 15, 1913, all in Folder 6:3, Box 6, Cox, DPL.

62. Claude E. Cox to John G. Callan, January 15, 1913, Box 6, Cox, DPL. Sloan noted that two years after its formation in 1908, General Motors absorbed twenty-five new firms. Alfred P. Sloan, Jr., My Years with General Motors, ed. John McDonald with Catherine Stevens (Garden City, N.J.: Doubleday & Company, 1964), 5, 7–9.

63. Efforts by Ford’s management to control automobile repairs are examined in Stephen L. McIntyre, “The Failure of Fordism: Reform of the Automobile Repair Industry, 1913–1940,” Technology and Culture 41 (April 2000): 269–99.

64. W. Benton Crisp to Mr. Plantiff, May 26, 1906, Box 37, Henry Ford Office Papers, Accession 2, Benson Ford Research Center, The Henry Ford (formerly, the Henry Ford Museum and Greenfield Village), Dearborn, Mich. (hereafter HF).

65. Settlement between Morton E. Duncan and the Ford Motor Company, November 23, 1905, Folder “General Legal Records—Claim Releases—Damage Suits—1903–1926,” Box 1, Office of the Treasurer—General, 1903–1933, Accession 483, HF; Directors’ Minutes, 82, November 17, 1905, Book #1, Box 1, Ford Motor Company Minute Books, Accession 85, HF.

66. Beecroft v. Van Schaick, 104 N.Y.S. 458 (1907). C. P. Berry, A Treatise on the Law Relating to Automobiles (Chicago: Callaghan, 1909), 209. Two treatises devoted to automobiles charted the growing variety and number of cases. See C. P. Berry, The Law of Automobiles, 4th ed. (Chicago: Callaghan and Company, 1924); and Xenophon P. Huddy, The Law of Automobiles, 5th ed. by Arthur F. Curtis (Albany, N.Y.: Matthew Bender & Co., 1919).

67. Greene v. Curtis Automobile Company, 129 N.W. 410 (Wis. 1911).

68. Ford Motor Company v. Morris R. Osburn, 140 Ill. App. 633 (1908). Other cases followed from a car’s defects and a buyer’s efforts to recover the price paid for a new or used vehicle. See, for example, Pitcher v. Webber, 68 A. 593 (Me. 1907); Boulware v. Victor Automobile Manufacturing Company, 134 S.W. 7 (Mo. Ct. App. 1911); Bedford v. The Hol-Tan Company, 128 N.Y.S. 578 (1911); and Klock v. Newbury, 114 P. 1032 (Wash. 1911).

69. Plaintiff’s Bill of Exceptions, 1–2, and Brief for Plaintiff, 1, Neale v. American Electric Vehicle Company, 186 Mass. 303 (1904), Case No. 2411, Massachusetts Reports Papers and Briefs, Volume 186, Supreme Judicial Court of Massachusetts, Suffolk, Social Law Library, Boston, Mass. (hereafter Records and Briefs for Neale).

70. Bill of Complaint, 1–3, Master’s Report, 16–26, Plaintiff’s Objections to the Master’s Report, 29–30, Masters & another v. Wayne Automobile Company & others, 198 Mass. 25 (1908), Massachusetts Reports Papers and Briefs, Volume 198, Supreme Judicial Court of Massachusetts, Suffolk, Social Law Library, Boston, Mass. (hereafter Records and Briefs for Masters).

71. Testimony of Harmon J. Hunt, 63–64, 77–81, 103–10, Testimony of David D. Buick, 65–66, Exhibit 79, The Irgens Auto Co. to Reid Manufacturing Company, June 16, 1904, 225–28, Exhibit 97, 236–38, Buick Motor Car Co. v. Reid Mfg. Co., 150 Mich. 118 (1907), Records of the Supreme Court of Michigan, University of Michigan Law Library, Ann Arbor, Mich. (hereafter Records and Briefs for Buick).

72. American Motor Car Co. to White Company, April 16, 1908, Exhibit A, 26, White Company v. American Motor-Car Co., 11 Ga. App. 285 (1912), Court of Appeals of Georgia, Case No. 4159, Box 80, The Georgia Archives, Morrow, Ga.

73. Alanson P. Brush, testimony, 350–51, quote 351, Records and Briefs for MacPherson.

74. Ford Motor Co. v. Livesay, 160 P. 901 (Okla. 1916).

75. Glenn Porter and Harold C. Livesay, Merchants and Manufacturers: Studies in the Changing Structure of Nineteenth-Century Marketing (Baltimore: Johns Hopkins University Press, 1971).

76. Ibid.; Chandler, The Visible Hand, 302–14, 403, quote 308–9. Harold C. Passer, “Development of Large-Scale Organization: Electrical Manufacturing Around 1900,” Journal of Economic History 12 (1952): 378–95; and Michael Massouh, “Technological and Managerial Innovation: The Johnson Company, 1883–1898,” Business History Review 50 (1976): 46–68.

77. Charles McCurdy, “American Law and the Marketing Structure of the Large Corporation, 1875–1890,” Journal of Economic History 38 (1978): 636–37. See as well, Chandler, The Visible Hand, 303–7; and Dicke, Franchising in America, 32–45.

78. Looking at distribution systems, Chandler did not distinguish between franchises or salaried agents. The examples I noted were cases where firms relied on salaried agents. I cited Chandler because he remains the most influential business historian. Other automobile historians did not pursue the question of liability except for Epstein’s brief note. See for example E. D. Kennedy, The Automobile Industry: The Coming of Age of Capitalism’s Favorite Child (1941; reprint, Clifton, N.J.: Augustus M. Kelley, 1972), 62–64, 127, 138–45; Rae, American Automobile Manufacturers, 46–47, 136–38; and Epstein, The Automobile Industry, 140, and in general 136–52. Dicke, relying on Hewitt’s study, noted the implications of liability for Ford. Hewitt misread Joslyn v. Cadillac, 177 F. 863 (6th Cir. 1910). The plaintiff (car buyer) contended that the dealer made false statements only in the sense that the product, being so defective, failed to live up to its advertised representations. Dicke, Franchising in America, 66–67; and Hewitt, Automobile Franchise Agreements, 19 n. 27, 24 n. 4, 25, 37–38, 45. I review Hewitt’s study in the Appendix.

79. Kessler, “Automobile Dealer Franchises,” 1135–36.

80. At least by 1912 the courts had established two features important for separating the agency and sales contracts. The Illinois Appeals Court heard a case in which the dealer claimed his 1909 contract constituted an agency relationship. Even when called an “Agency Agreement,” and even though the dealer was given an exclusive sales territory, the court declared that most parts of the contract were “wholly inconsistent with, and repugnant to, any theory of agency.” Vincent Bendix was prohibited from selling cars in the “name of the defendant” and had to pay for the cars “cash on delivery.” Bendix v. Staver Carriage Company, 174 Ill. App. 589 (1912), 595. See also Banker Brothers Company v. Commonwealth of Pennsylvania, 222 U.S. 210 (1911). The many cases regarding sales and agency contracts are reviewed in Berry, The Law of Automobiles, 1365–89, especially 1367, and Huddy, The Law of Automobiles, 1001–21, especially 1002–3. Although Hewitt did not review these two cases, he discussed many legal issues surrounding dealer contracts in Automobile Franchise Agreements.

81. Johnson v. Cadillac Motor Car Co., 261 F. 878 (2nd Cir. 1919).

82. Table 1 provides information about the firms’ production (Appendix).

83. The managers were paid a salary, but purchased cars at a discount as did other dealers. See, for example, Memorandum of Agreement between Thomas J. Hay and the Ford Motor Company, October 11, 1906, Folder 1, Box 1, Secretary’s Office—Contracts and Agreements, Accession 140 (hereafter Acc. 140), HF. Hewitt described other sales arrangements during Ford’s early years in Automobile Franchise Agreements, 19. See also Nevins, Ford: The Times, the Man, the Company, 264–65; and the Appendix.

84. W. E. Metzger to W. J. Stewart, December 18, 1902, Exhibit 5, 10–11, Wheaton v. Cadillac Automobile Co., 143 Mich. 21 (1906), Records of the Supreme Court of Michigan, University of Michigan Law Library, Ann Arbor, Mich. (hereafter Records and Briefs for Wheaton).

85. Table 2 indicates whether a firm included a clause in its dealer agreement declaring that the dealer was not its agent (Appendix).

86. Nevins, Ford: The Times, the Man, the Company, 264–65.

87. Leslie J. Stegh, “Putting America in the Driver’s Seat: The Deere-Clark Motor Car Company,” Illinois Historical Journal 81 (Winter 1988): 242–54; and Susan Helper, “Strategy and Irreversibility in Supplier Relations: The Case of the U.S. Automobile Industry,” Business History Review 65 (1991): 781–84, especially 792–93.

88. This clause was similar to the non-delivery clause, which is discussed in Hewitt, Automobile Franchise Agreements, 42; Kessler, “Automobile Dealer Franchises,” 1147, 1149; and U.S. Federal Trade Commission, “Report on Motor Vehicle Industry,” 142. Information is reported in Table 2 (Appendix).

89. See cases cited above, Ford Motor Company v. Osburn, Joslyn v. Cadillac, Buick, Neale, Masters, Washburn, and Levis v. Pope Motor Car Company, 95 N.E. 815 (N.Y. 1911). Jain examined other types of agency relationships in her study, arguing that as early as 1907 the courts limited the ability of bystanders to receive compensation for personal injuries sustained in auto accidents. In my study of manufacturer-dealer relations, I find that prior to 1908 a number of problems prompted by an auto’s defects, not just personal injury, gave automakers reason to elect the sales contract over the agency contract. Managers also added clauses to sales contracts limiting their obligations to dealers. Jain, “‘Dangerous Instrumentality,'” 67–74.

90. Brief for Defendant, 1, 4, Records and Briefs for Neale.

91. Plaintiff’s brief, 5, Records and Briefs for Masters.

92. Wheaton v. Cadillac Automobile Co., 143 Mich. 21, 106 N.W. 399 (1906), Records and Briefs for Wheaton. See also Hewitt, Automobile Franchise Agreements, 39 n. 34.

93. Ford Motor Company v. Osburn.

94. Joslyn v. Cadillac. See also Hewitt, Automobile Franchise Agreements, 37. The court discussed the vehicle’s defects in its lengthy opinion.

95. Ford Motor Company 1904 agreement, File # 7222–68–2, Box 871, Federal Trade Commission, Bureau of Corporations, Record Group 122, National Archives, College Park, Md. Warranty information is reported in Table 3 (Appendix).

96. Cadillac Motor Car Company 1908 agreement, Memorandum of Agreement, between Cadillac Motor Car Company and the Utica Motor Car Co., July 22, 1908, Defendant’s Exhibit 2, 61–64, Cadillac Motor Car Co. v. Johnson, 221 F. 801 (1915), Transcript of Record, Case No. 157, Records of the U.S. Courts of Appeals, Record Group 276 (hereafter RG 276), National Archives, Northeast Region, New York, N.Y.; Oakland Motor Car Company 1908 agreement, Memorandum of Agreement between the Oakland Motor Car Company and the Indiana Automobile Co., September 16, 1908, Plaintiff’s Exhibit 1, 49–51, Oakland Motor Car Co. v. Indiana Automobile Co., 201 F. 499 (7th Cir. 1912), Transcript of Record, Case No. 1891, RG 276, National Archives, Great Lakes Region, Chicago, Ill. For other firms, see Table 3 (Appendix).

97. Kessler, “Automobile Dealer Franchises,” 1141–42, 1147–48; and Hewitt, Automobile Franchise Agreements, 24, 31–32, 39. Macaulay, Law and the Balance of Power; Macneil, “Economic Analysis of Contractual Relations,” 1018–63; and Gordon, “Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law,” 569, 572.

98. Macneil, “Economic Analysis of Contractual Relations,” 1018–63; Kessler, “Automobile Dealer Franchises,” 1136. See also Gordon’s discussion of discrete transactions in “Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law,” 569–72.

99. Gen’l Attorney, Ford Motor Company, to William H. Atwell, August 27, 1913, Folder 75-29-24, Box 29, Ford Motor Company Legal Records, 1912–31, Accession 75 (hereafter Acc. 75), HF.

100. General Attorney, Ford Motor Company, to Mordecai & Gadsden & Rutledge, September 19, 1913; Mordecai & Gadsden & Rutledge to L. B. Robertson, telegram, September 26, 1913; Ford Motor Company to Mordecai & Gadsden & Rutledge, telegram, September 27, 1913; C. F. Rizer to Ford Motor Company, telegram, September 28, 1913; and Simeon Hyde to L. R. [sic] Robertson, November 26, 1913, File L-2020 Johnson, Daniel 1912–1914 (damage-personal), Box 3, Ford Motor Company Legal—Cases and Claims, Accession 297 (hereafter Acc. 297), HF.

101. Gen’l Attorney, Ford Motor Company, to W. K. Prudden Company, October 19, 1914, Folder 75-26-16, Box 26; and Gen’l Attorney, Ford Motor Company, to Jo Johnson, Att’y, April 20, 1916; Jo Johnson to L. R. [sic] Robertson, April 17, 1916, Folder 75-40-4, Box 40, all in Acc. 75, HF.

102. Ford wrote to its supplier, expecting the company would “reimburse us for at least fifty percent of the expense incurred in defense of this action.” Gen’l Attorney, Ford Motor Company, to W. K. Prudden Company, May 23, 1916, Folder 75-39-30, Box 39, Acc. 75, HF.

103. Lucy Shaffer was seated in a rumble seat bolted to a box on the rear of the car, and as the car climbed a hill the box cracked, causing the seat to give way. She was injured in the fall. Olds Motor Works v. Shaffer, 140 S.W. 1047 (Ky. Ct. App. 1911); Gen’l Attorney to Henry W. Thorne, March 1, 1915, File L-2008, Box 3, Acc. 297, HF.

104. General Attorney, Ford Motor Company, to Douglas B. Crane, Esq., June 6, 1914, Folder 75-26-16, Box 26, Acc. 75, HF.

105. Henry W. Thorne to Ford Motor Company, October 8, 1914, Folder L-2008, Box 3, Acc. 297, HF.

106. General Attorney, Ford Motor Company, to Douglas B. Crane, Esq., June 6, 1914, Folder 75-26-16, Box 26, Acc. 75, HF.

107. On autos, see Richard N. Langlois and Paul L. Robertson, “Explaining Vertical Integration: Lessons from the American Automobile Industry,” Journal of Economic History 49 (1989): 361–75; Michael Schwartz, “Markets, Networks, and the Rise of Chrysler in Old Detroit, 1920–1940,” Enterprise & Society 1 (2000): 63–99; and Benjamin Klein, “Vertical Integration as Organizational Ownership: The Fisher Body—General Motors Relationships Revisited,” in The Nature of the Firm: Origins, Evolution, and Development, ed. Oliver E. Williamson and Sidney G. Winter (Oxford: Oxford University Press, 1993), 213–26. See also Chandler, Scale and Scope; Oliver E. Williamson, “Transaction-Cost Economics: The Governance of Contractual Relations,” Journal of Law and Economics 22 (1979): 233–61; and idem, “The Modern Corporation: Origins, Evolution, Attributes,” Journal of Economic Literature 19 (1981): 1537–68.

108. Coase, “The Problem of Social Cost.” One of the few calls for a legal perspective in the study of the firm is found in Scott E. Masten, “A Legal Basis for the Firm,” in The Nature of the Firm, 196–212. See also Gillam, Products Liability in the Automobile Industry, 196–210; and McEvoy, The Fisherman’s Problem.

109. McEvoy, “The Triangle Shirtwaist Factory Fire of 1911,” 622–26, 643–48. See also Witt, “Speedy Fred Taylor,” 37–46.

110. Welke, Recasting American Liberty, 3–42, 112–24.

111. Lurie, “Lawyers, Judges, and Legal Change,” 31–56, especially, 43–49.

112. MacPherson’s accident is described in MacPherson (1912); Complaint, 4–6, Records and Briefs for MacPherson. See also Peck, Decision at Law, 41–42.

113. As a third defense unrelated to the question of defects, Buick claimed MacPherson drove too fast and his recklessness contributed to his accident and injuries. MacPherson (1914).

114. G. W. Durham, testimony, 178, quote, 185, Records and Briefs for MacPherson.

115. Charles Johnson, testimony, 231, quote 254, Records and Briefs for MacPherson.

116. On dealers and suppliers, see MacPherson (1916), 1053.

117. G. W. Durham, testimony, 189–90, Records and Briefs for MacPherson. See also Peck, Decision at Law, 47–52.

118. E. D. Cook, testimony, 208, Records and Briefs for MacPherson.

119. Each carriage maker began his testimony by giving his age, his years in the business, and his local residency. George A. Palmer, testimony, 50–57, Aldebert Payne, testimony, 57–72, and James P. Tittemore, testimony, 72–92, Records and Briefs for MacPherson.

120. George A. Palmer, testimony, 50–51, quote 51, Records and Briefs for MacPherson. See also Peck, Decision at Law, 47–48.

121. Payne declared: “I never let a wheel come painted; come oiled so I can see the quality of hickory used. I have invariably examined the wheels used in these various connections, with a view of determining the grades, and knots and defects.” Adelbert Payne, testimony, 56–59, quote 58–59, Records and Briefs for MacPherson. See also Peck, Decision at Law, 48.

122. James P. Tittemore, testimony, 72–74, Records and Briefs for MacPherson.

123. “Copy of Brief—Livesay Case,” no date, “Statement of the Issues,” 7, Folder 75-37-30, Box 37, Acc. 75, HF.

124. Respondent’s Brief, 59–67, Records and Briefs for MacPherson; Adelbert Payne, testimony, 62, Otto Kleinfelder, testimony, 92–94, quote 93, E. D. Cook, testimony, 208–9, Records and Briefs for MacPherson.

125. In his review of the testimony, Peck emphasized that Buick relied on the testimony of a civil engineer, W. K. Hatt, who had supervised the Forest Service’s tests of woods used for vehicles. According to Peck, Hatt discounted the carriage makers’ knowledge of wood, finding that the critical variable in assessing the quality of wood was a tree’s rate of growth as measured by the number of rings per inch. Growth of “five to twenty-five rings per inch” was considered, Peck noted, “good hickory.” Still, Peck omitted a key part of Hatt’s testimony. The engineer had agreed with the carriage makers in finding that weight, not just a tree’s rate of growth, was an important indicator of strength. His remarks echoed the Forest Service’s bulletin: “The best criterion of the value of the wood is its weight.” Moreover, while Hatt asserted the value of laboratory research, he sidestepped the question of MacPherson’s defective wheel. The 1908 study (circular 142, which Hatt cited) indicated that defects took many forms. In his 1933 study, Koehler cautioned against using a tree’s rate of growth as a criterion for assessing its strength. Peck, Decision at Law, 48, 50–51; W. K. Hatt, testimony, 101–35, especially 101–8, Records and Briefs for MacPherson; Anton T. Boisen and J. A. Newlin, “The Commercial Hickories,” Forest Service—Bulletin 80 (1910): 64; and Koehler, “Causes of Brashness in Wood,” 11–13, 18.

126. Nelson, The Legalist Reformation, 105–7, 187–88.

127. “Copy of Brief—Livesay Case,” no date, “Statement of the Issues,” 5–6, quote 9, Folder 75-37-30, Box 37, Acc. 75, HF; and “Points,” 2, Folder 75-37-30, Box 37, Acc. 75, HF.

128. Johnson v. Cadillac; Gen’l Attorney, Ford Motor Company, to Douglas B. Crane, Att’y, May 13, 1916, Folder 75-27-3, Box 27, Acc. 75, HF.

129. According to Gordon, Macaulay and Macneil recognized the importance of “mutual trust and solidarity” for economic development, but worried too that “power imbalances” could result in patterns of “persistent domination on one side and dependence on the other.” Gordon, “Macaulay, Macneil, and the Discovery of Solidarity and Power in Contract Law,” 570.

130. Helper, “Strategy and Irreversibility in Supplier Relations,” 792–99. See also Nevins, Ford: The Times, the Man, the Company, 220–51.

131. “Copy of Brief—Livesay Case,” no date, “Statement of the Issues,” 6–8, quote 8, Folder 75-37-30, Box 37, Acc. 75, HF; Ford Motor Co. v. Livesay, 901–3.

132. Johnson v. Cadillac. Jain concluded that MacPherson sustained a limited view of the automobile as “a mass-produced consumable and therefore did not press the issue of technical innovation.” In her reading, “the establishment of the National Highway Transportation Safety Administration (NHTSA) signaled a new common sense about the extent to which manufacturers should ensure that their products are crashworthy.” While important safety meansures dated to the 1960s and 1970s, I argue that a web of oversight agencies (public and private) did “press the issue of technical innovation” prior to 1960. The regulation did not focus on the vehicles’ overall crashworthiness, but did focus on critical components, such as brakes, bumpers, and safety glass. This is the topic of Part IV. I also argue that regulatory oversight was imperfect and managers made tradeoffs between safety and other concerns, such as the cost of production. Thus, autos housed some old defects and contained new ones as well. Jain, “‘Dangerous Instrumentality,'” 61–62, 75–77, quotes 62–63, 76. See also Eastman, Styling v. Safety; and Gillam, Products Liability in the Automobile Industry, 174–93. For a discussion of regulation in the years since 1960, see Jerry L. Mashaw and David L. Harfst, The Struggle for Auto Safety (Cambridge: Harvard University Press, 1990).

133. U.S. Federal Trade Commission, “Report on Motor Vehicle Industry,” 22; and Stanley Lebergott, Pursuing Happiness: American Consumers in the Twentieth Century (Princeton: Princeton University Press, 1993), 130.

134. More than two hundred manufacturers had sold cars in 1910, but by the 1930s the figure had shrunk to thirty according to Daniel Raff and M. Trajtenberg in “Quality-Adjusted Prices for the American Automobile Industry: 1906–1940,” in The Economics of New Goods, ed. T. F. Bresnahan and R. J. Gordon (Chicago: University of Chicago Press, 1997), 75. Laird noted the importance of repeat purchases in “‘The Car without a Single Weakness,'” 803–7.

135. Scott Knowles investigated product liability in terms of specific safety institutions, especially the Underwriters’ Laboratories, the Consumers Union and Consumers Research, Inc., which published Consumers’ Reports. Knowles, “Inventing Safety,” especially 184–290.

136. Walter P. Chrysler created his company out of the ashes of the Maxwell Motor Corporation. Flink, The Automobile Age, 70.

137. On the number of firms with large labs, see George Perazich and Philip Field, Industrial Research and Changing Technology, Works Progress Administration Report M-4 (Philadelphia, 1940), 66. These topics are discussed in Mowery and Rosenberg, Technology and the Pursuit of Economic Growth, especially 61–79. See also studies cited in note 23.

138. GM claimed seventeen percent of the market in 1911; its share dropped to twelve percent in 1921 and then climbed to thirty-four percent in 1929. U.S. Federal Trade Commission, “Report on Motor Vehicle Industry,” 22, 431. On the size of research staffs in 1927, see National Research Council, “Industrial Research Laboratories of the United States including Consulting Laboratories,” Bulletin 60 (1927), 23, 28, 48. In 1933, Ford listed one hundred researchers. National Research Council, “Industrial Research Laboratories of the United States including Consulting Laboratories,” Bulletin 91 (1933), 74. For 1938 data, see Perazich and Field, Industrial Research and Changing Technology, 66, 68.

139. On Kettering, see Leslie, Boss Kettering, especially 91–97, 183.

140. P. S. du Pont, President, to C. F. Kettering, July 19, 1921, File 87-11.7-61, Box 120; P. S. du Pont to C. F. Kettering, October 4, 1921, File 87-11.7-61, Box 120, Charles F. Kettering Archives (hereafter CFK), Richard P. Scharchburg Archives, Kettering University, Flint, Mich. (hereafter KU).

141. Leslie, Boss Kettering, 183–84.

142. Ibid., 197–99.

143. Alfred P. Sloan, Jr., Adventures of a White-Collar Man (New York: Doubleday, Doran, 1941), 160–63, 185–88; and U.S. Federal Trade Commission, “Report of the Motor Vehicle Industry,” 909–19. For a list of improvements, see G. M. Engineering Staff, “General Motors Developments,” December 10, 1956, File 87-4.19-16, Box 21, GMC/Proving Ground Collection, KU.

144. The Automobile Chamber of Commerce reported data collected by the National Safety Council on accident deaths beginning in 1913. Their figures show a decline from 306.7 deaths per 100,000 cars registered in 1913 to 105.9 deaths in 1926. During the 1930s, the number of deaths ranged from 111 to 144 per 100,000 cars. National Safety Council, Accident Facts, 1933 ed. (Chicago: National Safety Council, Inc., 1933), 34; and Automobile Manufacturers Association, Automobile Facts and Figures, 22nd ed. (1940), 84. By contrast, total accidents soared and were studied by legal realists in Committee to Study Compensation for Automobile Accidents, Report to the Columbia University Council for Research in the Social Sciences (Philadelphia: Press International Printing, 1932). Schlegel examined realists and social scientists in American Legal Realism, 105–14.

145. “Maker Liable for Car’s Defects,” Eastern Underwriter 17 (April 7, 1916): 1, 16; “Maker Loses New ‘Wheel Case,'” Horseless Age 37 (April 1, 1916): 289; Elton J. Buckley, “Manufacturer Legally Responsible for Damage,” Metal Worker 85 (June 23, 1916): 863; and A. L. H. Street, “Liability of Machinery Manufacturers,” Iron Age (January 21, 1915): 222–23.

146. Gillam described the standard warranty, but observed that in all the cases he reviewed automakers had never based their defense on the standard warranty’s disclaimer. Instead, they operated within the framework set out by MacPherson. Gillam, Products Liability in the Automobile Industry, 174–76, 189–93, quotes 162, 191. As I report in the Appendix, many manufacturers had included these warranties since the early 1900s.

147. Gillam cited the case of wheels to illustrate his point, noting that Ford had emphasized nine tests in its 1916 lawsuit. Gillam, Products Liability in the Automobile Industry, 188.

148. E. R. Jacobi, Kelsey-Hayes Wheel Corporation to Nicoll, Anabel & Nicoll, November 16, 1927; Hayes Wheel Company, “Testing Laboratory Some of the Standard Tests” (no date, pamphlet); [illegible] Cahill “Manufacture and Inspection,” November 16, 1927; Edward J. Foley, Wood Wheel Division, Kelsey Wheel Company, “Timber—Source of Supply, Method of Assembling Ford Automobiles,” November 15, 1927; “Torque Tests on Ford Wheels,” (no date [1929]); “A Comparison of Black Primers for Wood Wheels,” Hayes Wheel Company, Laboratory Report No. 3677, April 15, 1924; Laboratory Report No. 3945-6-7, May 14, 1924; Laboratory Report No. 3954 & 3955, June 5, 1924, all in Folder 75-78-5, Box 78, Acc. 75, HF. Gillam makes a similar point in Products Liability in the Automobile Industry, 188.

149. Gillam, Products Liability in the Automobile Industry, 188.

150. H. G. Weaver, “Tremendous Trifles,” October 12, 1932, 1–2, File 87-11.4-5; Customer Research Staff, General Motors, “Final Report of 1933 Surveys,” November 1933, 29, File 87-11.4-7, Box 110, CFK, KU.

151. Henry M. Crane, “More and Better Friends of General Motors,” Third General Motors Executive Conference, White Sulphur Springs, West Virginia, October 19–21, 1934, 8, File 77-7.4-1.13-1, Charles Stewart Mott Papers, KU.

152. “Product Meeting Held in Styling Auditorium on Thursday, October 16, 1941,” 2–3, File 87-11.4-19, Box 110, CFK, KU.

153. In the case of railroads, Welke wrote: “Technological advances or even simple operating changes could dramatically increase danger or introduce new hazards.” Welke, Recasting American Liberty, 28. Knowles traced the efforts of the Underwriters’ Laboratories and the Consumers Union to reduce new products’ risks. Knowles, “Inventing Safety,” 260–86.

154. Arthur T. Upson and Leyden N. Ericksen, “Wood Requirements of the Body Industry,” Automotive Manufacturer 65 (September 1923): 21–24, 29, quote 23.

155. The test was conducted in 1933. “Hand-Cranking and Carbon Monoxide,” Travelers Standard 24 (January 1936): quote 3–5. See also “Carbon Monoxide in Automobiles,” Travelers Standard 22 (March 1934): 41–52; and Eastman, Styling vs. Safety, 68–69, 177–78.

156. Cadillac and LaSalle were two exceptions. Eastman, Styling vs. Safety, 179–80. For a list of cars using safety glass, see A. M. Wibel to Edsel B. Ford, October 31, 1930, Folder, “Triplex Glass Data Working Papers,” MM-42-G, Box 96, Purchasing—A. M. Wibel, Accession 390 (hereafter Acc. 390), HF. Sloan wrote to Lammot du Pont, “I am trying to protect the interests of the stockholders of General Motors and the Corporation’s operating position—it is not my responsibility to sell safety glass.” Alfred P. Sloan, Jr. to Lammot du Pont, April 15, 1932, Defendants’ Exhibit No. 353, Deposition, Deft’s Exs. #18.3, volume 18, United States of America v. E. I. Du Pont de Nemours and Company, General Motors Corporation, United States Rubber Company, Christiana Securities Company, Delaware Realty and Investment Corporation, Pierre S. du Pont, Lammot du Pont, Irene du Pont, defendants, Civil Action No. 49–C-1017–1, Imprints Department, Hagley Museum and Library, Wilmington, Del.

157. Eastman, Styling v. Safety, 178–79.

158. “Notes on the Adoption and Use of Laminated Glass by the FMC,” April 12, 1932, Folder, “Data held for Pittsburgh Pate [sic] vs. Triplex,” Box 96, Acc. 390, HF.

159. Knowles argued that during the 1920s the UL acquired considerable scope and became a “‘national laboratory.'” Knowles, “Inventing Safety,” 195–211, quote 197, 219–59; and idem, “Lessons in the Rubble: The World Trade Center and the History of Disaster Investigations in the United States,” History and Technology 19 (2003): 9–28, especially 23–24. The UL was created in 1901 as a nonprofit research organization under the supervision of the National Board of Fire Underwriters, and its activities were described in “The Organization, Purpose and Methods of Underwriters’ Laboratories” (Underwriters’ Laboratories, 1921).

160. “Underwriters’ Laboratories Form Automobile Council,” S.A.E. Journal 3 (September 1918): 240–41; and A. R. Small, “Automobile Locking-Device, Classification and Theft Insurance,” S.A.E. Journal (August 1921): 97–99.

161. Sidney K. Bennett to Ford Motor Company, Attention Mr. Gaston Plantiff, July 11, 1921; W. T. Fishleigh, Experimental Engineer, to Underwriters’ Laboratories, attention—Mr. C. R. Alling, May 3, 1922; and W. P. Young to W. T. Fishleigh, May 8, 1922, all in Folder “Underwriters Lab. 1926,” Box 180, Engineering—Dearborn Laboratories, Accession 94 (hereafter Acc. 94), HF.

162. “Standards Committee Division Reports,” S.A.E. Journal 14 (June 1924): 628. See also Eastman, Styling vs. Safety, 12–13, 70–71.

163. “Substitutes for Ash in Automobile Bodies,” S.A.E. Journal 9 (September 1921): 181; and J. A. Newlin and Thomas R. C. Wilson, “Mechanical Properties of Woods Grown in the United States,” U.S. Department of Agriculture Bulletin 556 (1917).

164. American Instrument Company, “The James Brake Inspection Decelerometer” (Washington, D.C.: American Instrument Company, no date), in Folder “Brakes—1926, Ford, 4 Wheel, Misc.,” Box 166, Acc. 94, HF. On the Bureau of Standard’s role in developing brake equipment and standards, see “Brake-Performance Studies,” S.A.E. Journal 14 (February 1924): 236–38; and “Braking and Safety,” S.A.E. Journal 16 (January 1925): 19–21. The Bureau of Standards was created in 1901 and employed a staff of approximately 600 by 1917. Its work was described in “The National Bureau of Standards,” S.A.E. Journal 1 (August 1917): 132–39.

165. “Fire Hazards of Automobiles,” S.A.E. Journal 13 (September 1923): 247. Procedures for bumpers were described in Engineer, Casualty Department, to Manufacturers of Bumpers for Passenger Automobiles, May 16, 1924, Folder, “Underwriters Lab. 1926,” Box 180; and M. M. Brandon to Ford Motor Company, Attn: T. J. Little, Jr., November 28, 1925, and “Procedure for Inspections at Factories,” in Folder, “Underwriter’s Laboratories,” Box 159, Acc. 94, HF.

166. Underwriters’ Laboratories, “List of Inspected Automotive Appliances” (Underwriters’ Laboratories, April, 1921); and Underwriters’ Laboratories, “List of Inspected Automotive Appliances” (Underwriters’ Laboratories, April, 1922), Box 180, Acc. 94, HF.

167. “Improvements made on Ford Cars During the Last Three Years,” March 10, 1923, File 87-11.7-27, Box 119, CFK, KU.

168. W. T. Fishleigh to Mr. Liebold, May 2, 1925; Geo. D. Becker to Ford Motor Car Company, Attention of Mr. W. T. Fishleigh, September 9, 1925; W. T. Fishleigh, departmental communication, December 29, 1925; Geo. D. Becker to Ford Motor Car Company, Attention of Mr. W. T. Fishleigh, September 26, 1925; Geo. D. Becker to Ford Motor Company, Attention of Mr. W. T. Fishleigh, November 17, 1925, all in Folder “Underwriters Lab.,” Box 180, Acc. 94, HF. For other examples, see Knowles, “Inventing Safety,” ch. 6.

169. Nelson, “A History of the Forest Products Laboratory,” 159, 190–92, 205, 217–18.

170. Knowles found that the UL established the arbitration process in its effort to secure its authority in relation to clients. Disgruntled customers who had their goods rejected by UL engineers could appeal to the Bureau of Standards. Knowles, “Inventing Safety,” 241–46. On the SAE, see “Standards Committee Division Reports,” 625–28.

171. By 1921, fourteen states had passed laws for electric lights, and by 1925 forty-five had laws for electric lights. James J. Shanley, “Automobile Head-Lamp Regulation,” S.A.E. Journal 13 (September 1923): 249; and “Motor Vehicle Electric Lighting Laws,” Folder “Laws & Legislation 1926,” Box 172, Acc. 94, HF.

172. W. T. Fishleigh to Attention—Service Manager, June 3, 1922; Electrical Testing Laboratories, “Report No. 37168 Test of Ford Type H Automobile Headlight Lens Rendered to the Ford Motor Company, December 15, 1921; and Bureau of Standards, Department of Commerce, Report No. Tel-33025, “Test of 8-1/8 inch Type ‘H’ Ford Automobile Headlamps Submitted by State of New York, Tax Department, Albany, NY,” [stamped] January 21, 1922, all in Folder “Lights—Lens Ford Approvals,” Box 192, Acc. 94, HF.

173. “Improvements made on Ford Cars During the Last Three Years,” March 10, 1923, File 87–11.7–27, Box 119, CFK, KU. Eastman found that the problem of glare from headlamps for causing accidents had dated back to the 1910s. Eastman, Styling vs. Safety, 9–11, 13.

174. Manager, Washington Branch, to Attention Mr. Fishleigh, June 5, 1924; E. Austin Baughman, Secretary, Eastern Conference of Motor Vehicle Administrators, to Ford Motor Company, May 9, 1925; Frank A. Goodwin to the Automobile Headlamp Manufacturer, May 28, 1925; and Frank A. Goodwin, by A. W. Devine, “Bulletin Electric Headlamp Design and Construction,” November 18, 1925, all in Folder “Lights—Lens Ford Approvals,” Box 192, Acc. 94, HF.

175. Although I do not know the outcome of Hunt’s request, what I found important was the way he evaluated GM’s options. J. H. Hunt to C. F. Kettering, June 4, 1925, and J. H. Hunt to J. E. McEvoy, December 11, 1925, both in File 87–11.7–136, Box 121, CFK, KU.

176. Frank A. Goodwin to Ford Motor Co., November 18, 1922, and Frank A. Goodwin to The Manufacturers of Trucks, no date [stamped September 9, 1925 and September 18, 1925], both in Folder “Brakes—1926, Ford, 4 Wheel, Misc.,” Box 166, Acc. 94, HF. Eastman discussed safety problems for brakes, including the limitations of hydraulic brakes. Eastman, Styling v. Safety, 2–3, 12, especially 59–61.

177. Frank W. Judson to A. M. Wibel, November 12, 1931, with “Total Replies Received from Ford Owners—3955,” Folder “Pittsburgh Plate Glass,” Box 59, Acc. 390, HF.

178. Eastman, Styling v. Safety, 178–79.

179. Spence, “Consumer Misperceptions, Product Failure and Producer Liability,” 561. Welke offered an example of the first option in the public safety campaign for traveling on street cars and railroads during the Progressive era. Welke, Recasting American Liberty, 35–42. For an assessment of market failures by legal scholars, see for example, Croley and Hanson, “Rescuing the Revolution,” 683–797; and Hanson and Kysar, “Taking Behavioralism Seriously.”

180. Gillam, Products Liability in the Automobile Industry, 117–22, 126–28, 134–35.

181. Gillam noted manufacturers’ reliance on products liability insurance in ibid., 193–95.

182. L. B. Robertson, Gen’l Attorney In re: Julius Prezenik, Evelyn Prezenick, July 29, 1916, Folder 75-46-5, Box 46, Acc. 75, HF.

183. Gen’l Attorney, Ford Motor Company, to New England Casualty Company, January 20, 1915, and J. H. H. Liability Claim Department, New England Casualty Company, to L. B. Robertson, General Attorney, Ford Motor Company, January 23, 1915, Folder 75-30-6, Box 30, Acc. 75, HF.

184. See for example, Gen’l Attorney, Ford Motor Company, to New England Equitable Ins. Co., September 20, 1915, Folder 75-38-27, Box 38, Acc. 75, HF.

185. Claim Examiner to L. B. Robertson, Ford Motor Co., December 18, 1916, Folder 75-38-26, Box 38, Acc. 75, HF.

186. E. E. Juntunen, Legal Department, Ford Motor Company, to DeLancey Nicoll, Jr., December 12, 1924; John Vernon Bouvier, Jr. to De Lancey Nicoll, Jr., December 31, 1924; John Vernon Bouvier, Jr. to De Lancey Nicoll, Jr., November 19, 1927; and B. H. McNamara, General Accident Fire and Life Assurance Corporation, to Nicoll, Anable & Nicoll, December 22, 1927, all in Folder 75-78-4, Box 78, Acc. 75, HF.

187. John McTeer Briggs, “Products Liability Insurance” (Ph.D. diss., University of Wisconsin, 1953), quote ch. 1, 2; and M. A. Gesner, “Products Liability Insurance Is Constantly Growing Field; Prospects for This Cover Numerous,” Weekly Underwriter 132 (June 22, 1935): 1239.

188. James M. Cahill, “Product Public Liability Insurance,” Proceedings of the Casualty Actuarial Society 21 (New York: Casualty Actuarial Society, 1934–1935), 26; John M. Parker, “What About Product Liability?” Eastern Underwriter 40 (May 5, 1939): 27; and J. Harry Bibby, “How to Protect Yourself Against Lawsuits when Products Fail,” Sales Management 56 (June 1, 1946): 93.

189. “Big Auto Negligence Suit Against Ford Co.,” Eastern Underwriter 38 (January 8, 1937): 38.

190. Cahill, “Product Public Liability Insurance,” 26–29. Baxter v. Ford Motor Company et. al., 12 P.2d 409 (Wash. 1932).

191. Cahill, “Product Public Liability Insurance,” 28–30; “Demand for Products Liability Constantly Increasing with Many Potential Risks Uncovered,” Weekly Underwriter 140 (June 24, 1939): 1325–26. On the Uniform Sales Act, see also “Products Liability Needs Vigilant Treatment,” National Underwriter 41 (June 17, 1937): 26.

192. This public awareness in all likelihood reflected changes in Americans’ attitudes toward corporations and safety that dated to the Progressive era, yet the trade press did not delve into the political or medical context surrounding consumers’ lawsuits. For public attitudes about safety, see Knowles, “Inventing Safety”; idem, “Lessons in the Rubble,” 9–28; McEvoy, “The Triangle Shirtwaist Factory Fire of 1911,” 621–51; and Welke, Recasting American Liberty. See also Witt, “Speedy Fred Taylor,” 1–49.

193. L. H. Wiggers, “Products Liability Insurance,” Insurance Series Ins. 3 (New York: American Management Association, 1931), 3.

194. “Products Liability Becoming Big Problem for Retailer,” National Underwriter 40 (June 25, 1936): 30.

195. “Need Hold Harmless Cover,” National Underwriter 40 (May 28, 1936): 32; and Cahill, “Product Public Liability Insurance,” 27. Briggs recalled that hold-harmless clauses reflected the “bargaining strength” of retailers, and that underwriters were “not friendly” toward the agreements. Briggs, “Products Liability Insurance,” ch. 6, 34–35.

196. “Products Liability Sales Is Stimulated by Chain Stores,” National Underwriter 39 (February 21, 1935): 35. On Connor, see “Products Liability Needs Vigilant Treatment,” 26; and E. S. Banks, “A. P. Connor of Phila. Is Nationally Known Products Liability Expert,” Eastern Underwriter 37 (January 10, 1936): 34.

197. John Parker listed among goods subject to chemical tests at Travelers “lipsticks, beer cans, flash bulbs, pop bottles, tear gas, face powders, flavoring extracts, fire extinguishers, liquors, soft drinks, false eyelashes, aspirin, canned foods, tooth pastes, rouge, kitchen utensils, electrical fixtures and candy.” Parker, “What About Product Liability?” 27. E. R. Granniss, “Chemistry and Safety,” The Travelers Standard 21 (August 1933): 146–53. Briggs considered underwriters’ research studies standard in “Products Liability Insurance,” ch. 6, 2–25.

198. “Big Field Opened but It Needs Care,” National Underwriter 41 (April 1, 1937): 33, quote 48; and Briggs, “Products Liability Insurance,” ch. 6, 26–27.

199. See, for instance, Naomi R. Lamoreaux, The Great Merger Movement in American Business (New York: Cambridge University Press, 1987); and Thomas K. McCraw, Prophets of Regulation: Charles Francis Adams, Louis D. Brandeis, James M. Landis, Alfred E. Kahn (Cambridge: Harvard University Press, 1984).

200. Kessler, “Automobile Dealer Franchises,” 1135.

201. See for example, Stan Luger, Corporate Power, American Democracy, and the Automobile Industry (New York: Cambridge University Press, 2000).

202. Richard Swedberg, “Introduction to the Transaction Edition,” in Essays: On Entrepreneurs, Innovations, Business Cycles, and the Evolution of Capitalism, ed. Richard V. Clemence (New Brunswick, N.J.: Transaction Publishers, 1989), vii–xxxix. In closing, it is worth noting that, in wanting to explain why accidents mattered to Progressives, McEvoy, Welke, and Witt located cultural ideas about causation, corporate responsibility, and risk-taking. While agreeing with their findings, I studied technological defects associated with a new market. My project follows in the tradition of legal realists’ empirical research; my interests are seen as well in “Closing the Deal: GM’s Marketing Dilemma and Its Franchised Dealers, 1921–41,” Business History 45 (2003): 60–79. Valuable comments about realists may be found in Daniel R. Ernst, “The Critical Tradition in the Writing of American Legal History,” Yale Law Journal 102 (1993): 1019–76.




By: Sally H. Clarke



Are you passionate about history?

Share that passion with the world. Join us: Contribute to History Cooperative