History 2: The Collapse of Oligopoly
The End of Oligopoly
The durability of the Detroit Three oligopoly, given its high profitability, hinged upon the persistence of economies of scale. A number of smaller manufacturers did survive into the 1950s, including Studebaker, Packard, Nash, and Hudson. Two firms attempted entry, a de novo venture by Kaiser together with the moribund pre-World War II (WWII) producer, Graham-Paige, and Willys-Overland, which had made cars in the 1930s and the Jeep during WWII. Neither lasted long. Studebaker and Packard both exited, while Nash and Hudson merged to form American Motors (AMC) in 1954. AMC added jeep production in 1970, allied with Renault from 1978, and then was acquired by Chrysler in 1987 when Renault’s core operations in France ran into difficulties. Would-be new entrants proved unsuccessful in achieving a scale that would let them finance the development of new vehicles and keep engine and stamping plants sufficiently utilized to drive down average costs. There seemed to be room for only 3 significant producers in the US and Canada.
This was amplified by the focus of these smaller firms on compact vehicles. Costs change little with size: the components are the same and the assembly process similar. In the volume segments, the absolute price of cars increases with size. So for the firms on the margin, total revenue and per-vehicle profits were lower. Sharing key components with other firms to lower costs also proved impossible. For a variety of strategic and practical reasons, most car producers made their own engines, transmissions, and components where economies of scale were particularly important. The Detroit Three had no interest in selling to outsiders. The smaller outside firms sought to share engines, but products, product cycles, and capacity varied enough that this did not work out. Entry thus entailed not only designing and engineering a vehicle, setting up a dealership network, and building an assembly plant, but also building engine and transmission plants.
Nevertheless cracks in the oligopoly were apparent from early on. First, small cars remained a niche in the U.S. market, but one that expanded periodically. Smaller cars came into fashion every five or so years, for example in 1958, 1963, and 1968. Some of this demand was met by imports, particularly in the late 1950s, while in the late 1960s Volkswagen was able to sell 600,000 Beetles. The problem was with the Detroit Three. As an oligopoly, no member was willing to cede a market segment uncontested, and when demand for compact cars expanded, all launched downsized models to compete against AMC and imports. But for less expensive cars sales volume is crucial. By splitting the market none achieved enough sales for these cars to be profitable. However, it did suggest that the oligopoly remained vulnerable to entry from below.
Second, the comfort of oligopoly profits and stable market shares invited sloppy management. By 1970, finance came to dominate the executive suites of the Detroit Three, as at many other large U.S. corporations. In a stable competitive environment, the similarity of the players meant that new product ideas or efficiencies provided at best a fleeting advantage. Corporate politics became central to promotion. Managing factories was not a route to senior management, while turf wars resulted in a disconnect between styling, product engineering, purchasing, and manufacturing. Bland styles designed by committees entered production. Engineering was not involved in the initial stage so had to deal with styles that were hard to turn into vehicles, and the same problem affected the factories, as the design engineers were separate from manufacturing engineers and plant managers. The mandate of purchasing was to shave pennies off of prices, rather than to work with other functions to look for alternative materials and engineering solutions that would lower costs. Finally, as long as the Detroit Three were similar, high warranty costs were not treated as a symptom of problems that demanded attention. Executives could and did blame workers, not their own push for volumes and inattention to engineering quality into their vehicles. Above all, as long as all had equally poor quality, the oligopoly mindset meant that it was not a problem for management, only for consumers.
Third, there was the golden handshake with the union, since a boost of the UAW pay-scale meant a comparable rise for everyone else in the company. Higher pay was of course important to workers, but over time the unions sought and achieved greater job security, making their compensation packages less like an hourly wage and more like the annual salaries enjoyed by white-collar workers. This structure proved unsustainable when companies were forced to downsize, as the Detroit Three were still obligated to cover the benefits of retirees. From an economics perspective, the net effect was to turn labor from a variable into a fixed cost. Lower marginal costs, however, make it much more challenging for a cartel to maintain discipline. After 1975, the U.S. industry faced pervasive discounting.
The Oligopoly Collapses
Imports proved the straw that broke the Detroit Three oligopoly’s back. The compact car segment expanded with the quadrupling of oil prices that accompanied the first oil crisis in 1973. Without good domestic products, the share of imports rose, mostly from Japanese firms. Unlike in the United States, consolidation in Japan occurred much later, with nine producers of passenger cars still in the market in 1966. While GM and Ford had had plants in Japan from the 1920s, they chose not to restart operations after the war, viewing Japan as unlikely to ever arise from the ashes of defeat to comprise a significant market. The 30 or so Japanese firms that entered the market knew that they were high in cost and poor in quality, and attacking these challenges was a strategic focus: a firm that failed to lower costs and improve quality would soon be forced out of the market. They also had a hard time competing against imports, but the high tariffs imposed in 1956 were targeted for eventual elimination. This provided another incentive to improve the production side of their business. As the economy grew, expanding fast enough to meet demand was an additional challenge, again leading management to focus on the production side of the business. The bottom line was that by the early 1970s Japanese producers had largely caught up to U.S. standards, but kept improving their quality and productivity through better attention to their factories, one component of which was the “just-in-time” system we describe in Chapter 10.
By the mid-1970s, Japanese cars had attained a good reputation and begun building dealer networks. The Detroit Three believed that the small car segment would prove transient, but their own compact cars fared poorly against front-wheel-drive Japanese imports and did not sell well. As an alternative, they sought Japanese partners, with Chrysler taking a stake in Mitsubishi Motors, Ford in Mazda, and General Motors in Isuzu. Unlike in the 1950s and 1960s, however, the contrast in quality between Japanese products and those of the Detroit Three attracted a loyal following, and import sales did not fall when gasoline prices declined in the late 1970s. The market for compact cars then exploded with the second oil crisis, reaching a full one-third of the market. Japanese imports hit 1.68 million units before being capped by a quota, the voluntary export restraint (VER) negotiated by the new administration of Ronald Reagan.
The Detroit Three were correct in their belief that Americans would revert to buying larger vehicles, and by the end of the 1980s the share of compact cars was back into single digits. The VER was designed to give them breathing space. But trade restraints also encouraged Japanese car companies to build factories in the United States, beginning with Honda in 1982, Nissan in 1983, and Toyota in 1984. Even if virtually all parts were imported as knocked-down “kits,” the vehicles these factories made did not count against the VER. A second effect was that in order to limit exports, the VER effectively mandated that the Japanese companies form a cartel, allowing them to raise prices to the benefit of their bottom line. Third, when faced with a quota firms moved up-market. This provided an incentive for Japanese companies to invest in making mid-sized cars, which were a relatively unimportant segment in their home market. But between having factories in the United States and an array of larger vehicles, it meant that they did not fade from the market when sales of compact cars fell. Instead they were positioned with a range of price-competitive, high-quality cars that increasingly competed directly with the core products of the Detroit Three.
The gap in product quality persisted for almost 20 years, as detailed in the influential The Machine that Changes the World, the 1991 report of the International Motor Vehicle Program at MIT that documented in unambiguous terms the gap in quality and productivity between the Detroit Three and Japanese producers, and a smaller gap between German producers and the Japanese. It took roughly a quarter century, but eventually the Detroit Three came to operate with competitive levels of cost and quality. In the interim, the market share of General Motors fell from 50 to 17 percent, whereas the Detroit Three as a whole dropped from 85 percent to an average of 46 percent during 2010 through 2015. International trade and local production allowed firms to achieve economies of scale on a global basis, selling a small set of essentially identical vehicles in multiple markets. They were thereby able to spread costs without needing to have high sales in each and every market. But it is the bottom line that is key: the oligopoly was broken.
In Japan the consolidation of the market was delayed by the profits generated in the 1980s and 1990s by exports to the United States. Over time, however, several firms ran into difficulties, and Toyota gained control of Daihatsu, Hino, Isuzu and more recently took a large stake in Subaru and a smaller one in Suzuki. Excluding the latter, Toyota and its affiliates collectively control 50 percent of the passenger car market. Meanwhile Nissan entered an alliance-cum-merger with the French firm, Renault, and in 2016 agreed to take effective control of Mitsubishi Motors. Their combined share, however, will be only slightly over 10 percent. Meanwhile, Japan’s overall population is in decline, and due to aging the number of licensed drivers is falling even faster. Imports, primarily of German brands, account for more than one in six units sold in the market for full-sized cars. But with a declining market, of which an increased share is accounted for by low-priced minicars, the challenge is excessive capacity and struggling dealers. New entry is unlikely, but neither are high oligopoly profits.
Europe has seen a similar transition, as national producers were able to sell into a pan-European market. The inclusion of Eastern European markets after 1989 accelerated that process, as did the introduction of the Euro in December 1995, which removed foreign exchange risk. In addition, there was new entry into production within Europe by Toyota, Nissan, Honda, Suzuki, and the Korean Hyundai. This was paralleled by exit. In England, long-standing domestic brands such as Jaguar and Land Rover had already disappeared or been purchased by others, though Ford and Vauxhall (GM), both of which had entered the market in the 1920s, remained. In Spain SEAT was acquired by VW, as was Skoda in the Czech Republic. Renault bought the Romanian firm Dacia. Others, such as the maker of the East German Trabant, simply closed their doors. The bottom line remains the same: various national oligopolies disappeared, even as fringe firms exited.
Regulatory Change and Market Structure
The process of entry across borders via trade and direct investment has created a global market for vehicles. That was possible because selling the same model in multiple countries makes it easier to cover fixed costs. At the same time, engines and transmissions are easier to transport than finished vehicles and so the more capital-intensive parts of manufacturing can be concentrated in a smaller number of production facilities. Creation of a global fashion market also has created more commonality in the tastes of consumers for vehicle styling. Firms are thus able to design cars that sell well in multiple markets, something that was not the case in the 1980s. They can tweak seats and other parts of the interior, or offer manual transmissions or diesel engines in one market and not another without needing to reengineer the car itself.
Yet differing incomes and particular regulatory histories continue to shape national markets. In Japan tax, vehicle registration and licensing have created a unique minicar segment. These vehicles are typically narrower and have smaller engines than even European-made subcompacts, and virtually all output is accounted for by domestic firms. Since 2010, sales averaged about 35 percent of all passenger cars, while high gas prices mean that another third of the market is comprised of compact cars. In Europe, high fuel prices mean that cars also tend to be systematically smaller than in NAFTA. Diesel fuel however is lightly taxed, and hence is much less expensive than gasoline while the nature of diesel engines means that such cars are inherently more fuel-efficient. In France, for example, as much as 80 percent of sales can be of diesel vehicles. In contrast, diesels are a vanishingly small slice of the market in Japan, whereas in China diesel fuel is of such low quality as to not be compatible with modern, low-emission engines. While better fuel is now available, sales of diesel cars remain minuscule.
Finally, there is the United States. Very low taxes make gasoline comparatively inexpensive, and vehicles are on average far larger than in Europe and Asia. While minivans, sport utility vehicles (SUVs) and crossover utility vehicles (CUVs) are now found around the world, they were long distinctly American products. Similarly, almost all full-sized pickup trucks are sold in the United States. That is amplified by two other institutional features. One is the “chicken war tariff” of 25 percent on trucks, originally intended in 1963 as a temporary retaliatory measure against European restrictions on imports of frozen chicken from the United States. While defined broadly, in 1963 the VW bus of hippy fame was the only passenger “truck” imported by the United States. Over 50 years later that “temporary” tariff is still in place, and no longer innocuous as minivans and SUVs were classified as trucks. Until the last decade such vehicles remained the exclusive province of the Detroit Three.
Then there is CAFE, the Corporate Average Fuel Economy ceiling legislated in 1975, following the first oil crisis. It imposed steep per-vehicle fines on any manufacturer whose average car in a given year exceeded the legislated threshold. There was however an additional twist: imports were averaged separately from domestically made vehicles, and trucks were treated less strictly than cars. In practice only firms such as Porsche and Mercedes ever paid fines. CAFE created a strong incentive for domestic manufacturers to transform larger vehicles from “cars” into “trucks.” Combined with the protection provided by the truck tariff, the Detroit Three maintained a lock-grip on the pickup truck market and long dominated those for minivans and SUVs. With limited competition, these segments remain highly profitable, and this helps explain the relative inattention of the Detroit Three to traditional passenger cars, at least in their home market.
New Technologies and Model Proliferation: The Rise of Monopolistic Competition
From the 1920s, General Motors in the United States and Daimler in Germany understood the benefits of sharing parts and components across vehicles to obtain both the benefits of economies of scale in production and the better margins allowed by product differentiation. That is easiest with a body-on-frame architecture, in which the body and chassis are manufactured separately, and “married” in the final assembly process. This in principle allows a variety of bodies to be mated with the same chassis: a pickup truck, a delivery van, a passenger van, and an SUV can look very different while sharing most of the costly components in a vehicle. For passenger cars, however, a “unitized body” or unibody weighs less but has better handling and safety. Unibody architectures became common in Europe in the 1930s, and in the United States from the end of the 1960s. They are now used for passenger cars globally, and for most truck-like minivans and smaller SUVs.
In a unibody the floor pan and engine cradle, vehicle sides and pillar, and roof supports are all designed as a single entity, to which the external sheet metal is then attached. Such vehicles are lighter as the upper body and to a lesser extent, the sheet metal, add rigidity and crash protection and so allows other parts of the structure to be less robust. Creating multiple versions, however, becomes more challenging. Lowering the roofline, for example, would shorten the pillars that go from the lower body to the roof and change their angles and weld points. That in turn can affect vibration, crash performance, and other vehicle characteristics. While straightforward, validating such changes takes time and money. Similarly, creating a visually distinctive vehicle that can be sold as a different brand may require tweaking the width, the wheel base, and the shape of the front and rear of a vehicle. Although the engine and transmission could still be shared, engineering costs escalate and the potential to share parts not visible to the consumer—and to assemble it in the same factory—decreases.
The ability to develop flexible platforms improved over time, with experience and particularly with the gradual development of computer-based engineering. This was enhanced by the ability of some firms to export or assemble a model in multiple countries. Reducing costs, however, has a natural consequence: through sharing a platform a firm can develop two versions of a car aimed at slightly different target purchasers, and hence sell more vehicles. Globally however that results in more and more models in the market. In the 1960s, there were perhaps 60 distinct vehicles available in the United States; today there are over 300.
At the end of the day, a firm has more models, each selling at lower volume. With all the major firms doing the same thing, per-model production volumes decline, and there is no net improvement in profitability. This is the classical prisoners’ dilemma: no firm can afford not to pursue a new niche, such as CUVs, a more car-like version of the truck-like SUV favored by the proverbial soccer mom in the United States, Europe, and Japan. But if everyone enters every niche, no firm gains sales, but all incur increased costs.
This can be approached with the simple textbook model of monopolistic competition: as more and more firms, or here models, appear in the market, volumes fall and with that profit margins. In the limiting case, economic profits fall to zero. Second, it highlights the fragility of profits to modest changes in volume: a slight decline in volume causes average costs to rise sharply. Finally, in a monopolistic market every producer has excess capacity. In the automotive case, firms have become clever at building multiple models from the same platform on the same assembly line, and speeding the ability to swap dies to stamp different doors on the same machinery. Balancing the production of more models, each at smaller volumes, makes the job of a plant manager challenging.
The overall dilemma of the virtual car, where computer-based engineering makes it possible to test a design without ever making a physical prototype, is that it makes earning a profit more difficult. Designing a new vehicle—or tweaking the styling of one that sells poorly—can now be done rapidly and at low cost, potentially in under 2 years when the changes are modest. Welding robots can be reprogrammed, and the level of training of assembly workers means a different version of the same platform can be brought into production without slowing output. Cutting the steel of the dies to stamp out body panels may be the most time-intensive part of the process. But all that means that a really successful design has only a 3-years window to earn above-normal profits before rivals begin to respond.
To date there are two main exceptions: pickup trucks and luxury car brands. The former is not entirely true because there has been new entry, with both Nissan and Toyota making full-sized pickup trucks in the United States. To date, however, their sales have not been sufficient to upset the dominance of the Detroit Three. However, entry has been able to partially eclipse the sales of jeep-like vehicles, as multiple firms now sell products that sell well in that segment. The success of pickup trucks is enhanced by a longer product cycle, which makes them more profitable for incumbents while making success by new entrants more challenging. The experience of the industry suggests the Detroit Three lead will erode over the next two product cycles, or 10 to 15 years. Indeed, this has already taken place in the market for jeep-like SUVs and CUVs.
Luxury marques succeed for a different reason: cars are status symbols, and brands such as BMW and Lexus sell on that basis. The improvement in engineering tools means that a Hyundai or a Chevy can be as quiet as a “luxury” brand, and offer an array of options that provide similar user comforts to a car nearly double the price. They do not however confer status. Breaking into that exclusive set of brands has proven difficult, and rebuilding a luxury brand that loses its mystique can require three product cycles of 4 to 5 years each and billions of dollars. Among others, the Lincoln brand of Ford and the Acura brand of Honda struggle to achieve anywhere near the volumes of a Mercedes. The challenge here is that for luxury brands to expand, they need to sell “down market” and smaller vehicles. Price competition becomes more important, and it becomes hard to claim that a smaller car represents a luxury purchase. Unless such vehicles can maintain a separate identity, as with BMW’s “Mini,” the profits that would come from success in the middle and lower end of the market can undermine those at the luxury end where these firms make their profits. The empirical question is whether there is a limit to the number of brands at any one time that can be perceived as “luxurious”. The greater that number, the more difficult it will be for these firms to maintain their profitability.