Consider, first, the upheaval in production processes. Within the past decade automation, robots, machining centers, and numerical controls have increased productivity significantly in the broadest sense. These innovations have reduced the work force, expanded output per unit of time, allowed quick changeover from one type of manufacturing process to another, and permitted greater flexibility in plant siting. Translated into quantitative terms, the labor content of traditional assembly operations has dropped from 25 percent to somewhere between 10 and 5 percent of the total cost of the product.
In turn, the microchip's increased memory and intelligence has precipitated a steep decline in the cost of computer power. Together with automated production processes, computer-aided design and computer-aided manufacturing are begetting upheavals of their own.
The competitive repercussions of this shift from labor to capital in production are already evident in the automobile industry. The Japanese automobile industry, as a whole, produces over 13 million units a year and employs only 670,000 people. This includes the work force of all 11 automakers, their component suppliers, and automobile contractors. In contrast, General Motors (GM) alone had a global work force of more than 690,000 persons at the end of 1983.
The cost of labor at Nissan is under 7 percent of total costs and it is less than 6 percent at Toyota. A direct comparison to U.S. labor costs is difficult because of the different ratio between components manufactured and purchased from other sources. But in fiscal 1981, when comparable data became available, the cost of labor for Ford (with an approximately 45 percent "make" ratio) was $10 billion; at General Motors (with a 70 percent "make" ratio), labor costs ran to $18 billion; at Toyota, the cost of labor was $1 billion. Even if Toyota's wage rates matched GM's dollar for dollar, Toyota's profit loss would be less than 1.1 percent from its recent 7 to 8 percent return on sales. The difference is not so much in the wage rate as in the labor content, that is, how many minutes does it take to produce a car?
Toyota, which produces 3.3 million units a year, has kept reducing production man-hours so as to maintain its employee level at about 45,000 persons during the past decade (prior to its merger with its sales company in 1983) while increasing its output 3.5 times. Nissan's productivity -- which is about twice that of its global competitors -- has taken the same route. These companies have changed the traditional labor-intensive auto industry into a capital-intensive industry.
The story is the same in electronics. During the past five years, the work force required to assemble a given consumer electronics product has been halved. Today, direct labor costs in this industry are, on average, down to 5 percent of total costs. Likewise, the semiconductor industry has become a fixed-cost, capital-intensive game, as opposed to the variable-cost, labor-intensive game of only half a decade ago.
The trend is even more prevalent in the continuous-processing industries -- chemicals, textiles, steel, and so forth. In these industries automated control systems increase productivity and hence competitiveness. In two of Japan's leading steel mills, Nippon Steel and Nippon Kokan KK, the labor tab hovers around 10 percent of total costs.
Expense of Cheap Labor
Probably the most far-reaching implication of this shift from labor to capital is that it shatters the mirage of low-cost labor in developing countries. The rationale behind companies locating their operations in countries where low labor costs were available was to focus management attention on bringing down variable costs. True, labor costs in developing nations still are only one-third as high as in developed nations. However, now that the direct labor content in competitive companies represents less than 10 percent of total manufacturing costs, the advantage gained by using cheap labor is offset by the costs of transporting critical components to the production site from developed countries (as most developing countries do not have the requisite vendors and subcontractors), and the costs of insuring the finished products and transporting them to major markets.
It costs more to bring products from cheap labor countries to the major markets in the industrialized nations belonging to the Organization for Economic Cooperation and Development (OECD) than to produce them close to home. For example, the typical cost of transporting a color television set from Southeast Asia to the West Coast of the United States is 13 percent of free on board (FOB), including duties and insurance. As a result, savings in the labor cost up to 10 percent of the total cost are offset by the additional transportation cost. On top of this, if critical components have to be imported to these countries from Japan and Taiwan, which is usually the case, the net result of the trade-off would favor siting the production facility in the market area or in the area where important component parts are available. This move also makes sense in industries where new models are constantly superseding old ones and hence there are constant changes in molds, jigs and tools, and components. Consumer and office electronics has become largely a fashion industry, where "booms" are normally short-lived, and certainly a production location remote from the location of the core engineering group has become very inconvenient. As a result, the attractiveness of producing goods in developing countries (relative to OECD) has almost disappeared. The situation is often compounded by the shortage of a good labor force and the lack of qualified local managers in developing countries.
This cost factor is why most blue chip Japanese companies no longer seek out the cheaper labor offered initially by Korea, Taiwan, and Singapore, and subsequently by Thailand, Malaysia, Indonesia, and the Philippines. It is also the reason why some American and European semiconductor manufacturers who kept chasing cheaper labor in these countries have fallen behind in the integrated circuit/LSI market share. Many firms set up shop in low-wage countries only to discover that their total costs did not drop significantly, or did not stay low for very long. Nippon Electric Company has withdrawn production, particularly the labor-intensive packaging and bonding processes, from scattered sites in Southeast Asia and consolidated its operations in a highly automated plant in Kumamoto, Kyushu. Hitachi and Toshiba have followed Nippon Electric. The chip-makers have learned first-hand what CTV and textile industries discovered earlier: inexperienced labor must be trained and, once trained and experienced, labor does not stay cheap very long. Therefore, those who succumb to the mirage of cheap labor constantly migrate in pursuit of it. Right now, they are somewhere in India, Sri Lanka, or Indonesia. It is likely that they will all end up in the People's Republic of China where one billion people have just joined the low-wage labor market in Asia.
Flexibility of Options
Managers in automated industries who fail to recognize the implications of this shift from labor to capital will find their companies squeezed between spiraling inflation and excessive labor costs. The strategic ramifications of this fundamental shift include an increased ability of automated operations to fight inflation. The reason is that the ratio of labor costs to total manufacturing costs is bound to increase when compared to declining sales or inflated wages. Automated operations also resist recession. For example, highly automated facilities such as Yamazaki (machine tools) and Fujitsu Fanuc (numerical controls) are said to break even at 10 percent capacity; others like Toyota (passenger cars) claim that they can operate at 70 percent and still not lose money.
The other side of this shift from labor-intensive to capital-intensive industry is that it demands deep and immediate market penetration. This is vital to ensure the maximum economies of scale needed to defray heavy initial investment and to sustain the heavy outlays necessary for continued production process innovation. Domestic markets, even given the size of Japan or the United States, have proven to be too small for global-class automated plants in semiconductor and machine tools.
Therefore, it is more critical than ever before to be close to the market in order to keep product lines readily accepted by the majority of the target customers and in tune with competitive demands. New products that anticipate or serve user needs and the use of strongly entrenched distribution channels to reach prime markets may be key success factors, especially once a product reaches a "commodity" status and can be made by numerous competitors. At that stage, the opportunity to reduce costs to establish a competitive edge in a highly automated industry is pretty much the same for all participants. Since it has become very difficult to keep the product's competitiveness through differentiated technologies and design, it is mandatory that the company maintain the distribution capability to win the nondifferentiated game of "engineered commodities." Hence, these commodities are no different from sugar and cement, in which product differentiation is extremely difficult. Although it is very difficult to learn how to make a high-quality color television, as many as 30 companies around the world have mastered the art and can each now make products that are virtually indistinguishable in terms of quality. As a result, a company's ability to sell large volumes of nondifferentiated products at the lowest cost to the end user has become the key factor for survival.
Copyright © 1985 by Kenichi Ohmae and McKinsey & Company, Inc.