Time to Get Serious About Workplace Change
Big productivity increases are possible for companies switching to high-performance work systems; government can aid the transition.
In the early 1990s, Lockheed Martin's Government Electronic Systems plant in Moorestown, New Jersey, was suffering from the decline in defense markets. Layoffs were widespread. By 1992, it looked as if the plant would have to shut down, eliminating hundreds of jobs. But management made an enlightened decision: It formed a joint partnership with the International Union of Electrical Workers Local 106 to implement a high-performance work system. The company stopped outsourcing subassemblies and ordered new technology. Workers, in cooperation with management, redesigned the work flow and created a new training program. In short order, they turned the plant around. Between 1992 and 1995, productivity increased 64 percent. Scrap and defects were reduced by over 80 percent. Product cycle time was cut by 50 percent, inventory by 80 percent, and manufacturing costs by 25 percent. Not only have jobs been saved, but the job loss was reversed.
In the midst of massive downsizings and plant closings, some of the United States's most forward-looking companies have transformed themselves with the aid of high-performance work systems: Corning, Folgers Coffee, Harley-Davidson, John Deere, LTV, Magma Copper, Mercury Marine, Reynolds Metals, Rockwell, Union Carbide, Weyerhauser, and Xerox, among others. Their success, bolstered by a growing body of research about the organizational importance of deep worker involvement, indicates that high-performance work systems are the best way to leverage the capabilities of a company's workers to achieve impressive gains in quality, productivity, and profits.
A high-performance work system seeks to enhance organizational performance by combining innovative work and management practices with reorganized work flows, advanced information systems, and new technologies. Most important, it builds on and develops the skills and abilities of frontline workers to achieve gains in speed, flexibility, productivity, and customer satisfaction.
Unfortunately, only a relatively few companies are pursuing this approach. Most chief executive officers and directors of large companies see high-performance work systems as risky because they require a sweeping change in operations. They find it easier to cut costs by laying people off. Although this may improve the bottom line for a few quarters, it does little or nothing to ignite growth, add jobs, or improve competitiveness. And small and mid-sized companies-the economic bulwark of small towns and the engine of this country's economic growth-simply don't have the knowledge, tools, or resources to implement these systems.
Neither the marketplace nor current public policy seems to be able to provide sufficient incentives for companies to develop high-performance work systems. Without this investment, however, U.S. companies will continue down the well-trodden low-road path, laying off more and more workers, outsourcing more work, and further weakening our nation's ability to compete. We cannot cut our way to jobs and growth.
With only modest changes in a variety of existing federal programs, government can play a role in countering this trend. The government can help break down the barriers to implementing high-performance work systems and speed the diffusion of this new form of work. As an enabler, the federal government can support the development and diffusion of training, tools, technologies, technical assistance, standards, and resources that will make it possible for companies to reap the benefits of this vital approach.
Ultimately, however, businesses and workers must take the lead in fostering change. Success will depend on how well the system performs in the competitive marketplace. But public policies can help companies overcome the initial hurdles. With minimal investment, the government can help preserve and expand jobs, bolster economic growth, and improve industrial competitiveness.
Worker involvement is key
In recent years, a majority of U.S. businesses have adopted one or more innovative work practices. These include quality circles, flexible job classifications, cross-functional training, pay-for-performance compensation systems, and various forms of employee involvement. However, very few have adopted the full complement of innovative work practices associated with a high-performance work system, which has three main components.
Worker participation. The key characteristic of a high-performance work system is extensive worker participation in all aspects of the company. Many companies already practice some form of worker participation, from suggestion boxes and quality circles to self-directed work teams. The extent of discretion given to workers in influencing and making decisions varies greatly. The payoffs are higher quality goods and services, improved workforce productivity, and greater company flexibility.
Studies indicate that most forms of worker involvement are an improvement over the traditional mass-production approach, in which workers perform manual tasks that require little thought and provide them with few opportunities to improve the process. The recent Commission on the Future of Worker-Management Relations (known as the Dunlop Commission) found that employee participation, when sustained over time and integrated with other organizational policies and practices, results in positive economic gains. A 1990 interindustry survey of 495 major businesses concerning their participation and employment practices, by Daniel Mitchell of the University of California at Los Angeles, David Lewin of Columbia University, and Edward Lawler III of the University of Southern California, concluded that the extensive participation of clerical and production workers led to significant improvements in return on investment, return on assets, and productivity. Other studies show that deep worker involvement is also essential to corporate flexibility and quick customer response; in today's volatile product markets, only the innovative and responsive survive.
In their 1994 book The New American Workplace, Eileen Appelbaum of the Economic Policy Institute and Rosemary Batt of Cornell University review a wide spectrum of research that shows that the share of firms with "at least one employee-involvement practice somewhere in the company is large and growing, and a significant number of firms have begun to make more extensive use of these practices." Although encouraging, these efforts fall short. What distinguishes a high-performance work system is deep participation and control by frontline workers. This requires redesigned machinery and software that allows substantial worker input, information systems technologies that help frontline workers coordinate these machines, and training that enables workers to use them. There is strong evidence that companies in many industries increasingly understand that they need these things but confront large hurdles in investing in their development. In addition, the experiences of a few large companies that are trying this approach can be misleading, because the development to date has tended to be piecemeal and incomplete, and consequently not always successful.
It should be pointed out that because of the importance of worker participation in achieving flexibility, companies in stable markets competing on price in the sale of commodity products may not benefit as much from a shift to a high-performance work system. In such cases, companies must carefully examine whether the productivity gains achieved by adopting a high-performance work system outweigh the cost of the transition.
The work system. Giving employees the depth of control and responsibility that is characteristic of a high-performance work system requires overhauling the institutional systems surrounding them as well as redesigning work flow. This will require the adoption of new types of production system layouts, improved communication that gives workers the information they need, and a commitment to continuous training that emphasizes not only job-specific technical skills but also process skills such as statistical quality control and learning skills such as problem solving.
Frontline employees, or their representatives, will also have to participate in important nonproduction functions, such as strategic planning, product design, customer and supplier relations, and equipment and technology decisions. Finally, the system must include a compensation structure, performance appraisal process, and other motivational techniques that reward participation and the taking of responsibility.
The most thorough appraisal of the need for a holistic approach was presented by Casey Ichniowski of Columbia University and researchers from four other universities in the July 1996 issue of Industrial Relations. They found that productivity gains are greatest when firms adopt "bundles" or systems of related innovative work practices to expand worker participation and flexibility in workplace design-the underlying premise of high-performance work organizations. A multiyear study of unionized steel-finishing lines by Ichniowski, Kathryn Shaw of Carnegie Mellon University, and Giovanna Prennushi of the World Bank demonstrates that the adoption of an integrated system of innovative practices substantially improves productivity and quality, whereas adoption of individual work practices has little or no effect. Studies of the automobile, apparel, electrical components, metalworking, and machining industries reach similar conclusions. In short, the whole system must be changed to significantly improve a company's performance.
Smart workers, smart machines. Strong evidence exists that worker involvement in designing work systems and choosing the production technology a company deploys enhances that company's flexibility and productivity. According to many organizational researchers, the decisive factor is not the technology per se but the complex interaction between workers and process technologies. Companies too often attempt a technological quick fix, such as introducing advanced computer-aided manufacturing equipment, without considering the kinds of work organization and practices, skills training, and compensation systems needed to achieve their objectives. As Harvard Business School professor David Upton wrote in a 1995 study of modernization in the paper industry, "Most managers put too much faith in machines and technology, and too little faith in the day-to-day management of people."
The evidence is strong enough to indicate that the alternative strategy of replacing workers with automation usually falls far short of expectations. University of Southern California management professor Paul Adler and Stanford University computer scientist Terry Winograd argue that if organizations fail to design work systems around people and with their input, new technologies will realize only a fraction of their potential benefit. In a 1994 Labor Department report, Integrating Technology with Workers in the New American Workplace, Scott Ralls, now director of economic development for the North Carolina Community College system, chronicles extensive research that indicates that organizations as well as workers benefit when firms invest in workers' technical training, involve workers in the continuous adaptation of technology to increase business effectiveness, and involve workers in the design and implementation of new technology in the workplace.
For example, a nationwide survey of 584 plants with metal-cutting machines concluded that plants in which all machine operators routinely wrote and edited the software programs that controlled the machines were 30 percent more efficient than plants where production workers did not. Another survey of 100 top executives found that employee involvement in the design and implementation of the company's information technology systems was one of the most common factors in the success of those systems.
At Sikorsky Aircraft in Stratford, Connecticut, machine-shop workers were directly involved or consulted in process changes and the selection of new equipment. As a result, operations shifted from the classical production line setup to "cell production," in which a small group of workers at one station performs all the operations needed to build an entire unit or subassembly. Line-operator input in the planning of the facility's layout, the evaluation and selection of new equipment, and the improvement of machining processes critical to better production efficiency was considered invaluable by management at all levels.
Barriers to high performance
Although high-performance work systems are gaining acceptance in business, progress has been agonizingly slow because many institutional, organizational, technical, and financial barriers must be overcome. Moving away from the traditional command-and-control model, where managers make decisions and workers take orders, is especially difficult. A successful transformation requires a substantial commitment of time, resources, and personnel. Work flows are disrupted. Managers and workers must be retrained. For many corporate leaders, closing a plant is much simpler than transforming it, despite the negative effects on the company's long-term profitability and growth potential.
Furthermore, there is still great reluctance to implement high-performance work systems because the transformation can go wrong in many ways. It is risky and potentially costly. Misguided efforts to put extensive robotics in manufacturing plants in the 1970s and 1980s, for example, hurt many firms. Creating a successful high-performance work system goes far beyond reengineering; it requires transforming a business culture. And in most cases, even managers, engineers, and workers who want to make the change don't have the required knowledge, experience, training, tools, technologies, financial resources, or incentives needed. A company that buys advanced computer-aided design and manufacturing systems, for example, will not achieve desired productivity and performance gains if it fails to produce advanced work practices, skills training, and compensation systems that enable employees to effectively use them.
Overcoming these problems requires a supportive environment and the creation of tools and techniques that are difficult to find today. For example, machine tools are still generally designed to minimize worker skills. Corporate information systems are still oriented toward supplying command-and-control information to management rather than providing production information to frontline workers. Technologies for involving frontline workers in product design are still in their infancy.
Market forces will not necessarily foster innovation, either. In fact, they may discourage it. Pressure by investors makes companies focus on the next quarter's earnings. Cost-cutting has become such a stampede that it's harder than ever for a company to make long-term investments. Shareholders say, "Everyone else is cutting costs. Why aren't you?"
The federal government obviously cannot jump-start the transformation by legislative or regulatory fiat. But it can serve as a catalyst and enabler. Government policy can help foster economic, political, and social environments that favor and speed the adoption of high-performance practices and reduce the risks and costs of implementation. The government can support the development and diffusion of tools, technologies, technical assistance, and standards that make it possible for companies to move toward high-performance work systems, and it can help expand the educational and training resources required.
Fostering high-performance work systems will not require an extensive revamping of federal R&D policy; rather, it will largely involve refocusing existing policy tools and programs. Although relatively little is being done now, many existing technology programs can be infused with the goals and priorities of high performance.
Basic research. To be able to develop the required manufacturing and information technologies, additional research on the basic science of high-performance systems is needed. Industry needs "soft" technologies that are associated with workplace change and skill development, such as workflow designs, and "hard" technologies such as computer interfaces and manufacturing machines.
Especially important is a redefinition of the relationship between workers and machines, focusing on how worker participation can be encouraged in the design of products and processes. Multidisciplinary research, both theoretical and empirical, is needed. Research with high-performance criteria in mind is particularly needed in traditional industrial engineering areas, such as plant layout and work floor design.
As the nation's leading basic research agency, the National Science Foundation (NSF) should create a new initiative in high-performance systems research that coordinates and expands existing work in its social, behavorial, organizational, and industrial sciences programs. For example, NSF's Transformation to Quality Organizations, Management of Technological Innovation, and Societal Dimensions of Engineering, Science, and Technology programs support some relevant research projects on workplace and organizational change or provide sufficient scope to support research on related topics. High-performance criteria and goals should also be injected, as appropriate, into NSF's traditional manufacturing, industrial engineering, and computing research programs.
Applied research. Applied research is needed to create the general hardware and software, process technologies, design tools, advanced systems, and devices for high-performance work systems. At the heart are systems that enable companies and their workers to keep in close touch with supplier and customer needs, and processes that give workers greater monitoring, control, and troubleshooting capabilities. Together these abilities make production more flexible and productive.
Three areas of "hard" technologies must be addressed. First are technologies that enhance workforce control over production, workplace organization, and machinery. This includes hardware and software that expand employee problem-solving, decisionmaking, and judgment capabilities at the point of production; skill-leveraging automation; and human-machine interfaces that enhance worker control and increase the level of knowledge and skills needed to program and operate advanced machines.
Equally important are technologies that enable workers to participate in integrated product design, development, and implementation. These include advanced computing and telecommunications technologies such as simulation, virtual reality, database and networking technologies, and the distribution of these on information networks. Progress is being made. For example, in designing the new 777 airliner, Boeing used a computer-aided design system that allowed design-and-build teams to work with mechanics and ground crewmen to evaluate how different options would facilitate future repair and maintenance. Further R&D should build on these successes.
In addition, education technologies, tools, and methodologies are needed to help the nation's managers, engineers, and workers obtain the skills required to operate in high-performance work environments. Examples include computer-based multimedia and advanced simulation software.
"Soft" technologies are needed too, such as benchmarking and best-practice assessment tools; new metrics and standards for evaluating high-performance practices; new methodologies, models, and metrics for designing, implementing, and evaluating high-performance transformations; participatory design methodologies and tools; skill assessment and development tools; worker-centered production, scheduling, and quality control methods such as statistical process control; and methodologies and tools for technical assistance providers.
Pursuing this high-performance research agenda should be a focus of the government's primary sponsors of applied research: the Defense Advanced Research Projects Agency, the Department of Energy (DOE), the National Aeronautics and Space Administration, and the Department of Commerce's National Institute of Standards and Technology (NIST). These agencies already sponsor important related work, but the emphasis on high-performance work systems should be strengthened and woven more tightly into their programs. For example, worker input into technology design should be given greater emphasis in NIST's Advanced Technology Program and the DOE Office of Industrial Technology's "Industry of the Future" projects.
To help reorient applied research, an explicit version of the government's "critical technologies list" should be established for high-performance work systems. Indeed, certain information and process technologies that are already on the critical technologies list are key to high performance; explicit recognition of this link would greatly encourage federal appropriations to existing programs that would speed development of high-performance systems.
The National Information Infrastructure (NII) projects should also be reexamined to ensure that they promote the worker-leveraging elements of high performance. NII programs in advanced computing, advanced networking, and telecommunications can provide much of the infrastructure backbone needed for high-performance systems. In addition, the definition of infrastructure that qualifies for federal and state economic development grants should be broadened to include advanced telecommunications technologies that expand high-performance systems.
The national laboratory system should also play a role. Cooperative research and development agreements (CRADAs) that allow companies and the labs to work together on advanced technologies could easily be extended to high-performance projects. A special fund should be set aside for CRADAs involving this work. Alternatively, the government could increase its support of cost-shared projects for high-performance systems.
Risk reduction. Federal technology programs often include demonstration projects and test-bed activities. Demonstration projects provide the opportunity to try new ideas and determine what public infrastructure is needed to support a new technology. Test beds allow technology to be pushed to its limits without the fear of failure or heavy financial losses by one company or industry. Both vehicles make possible the real-world experimentation necessary to move from concept to practical development.
Chief executive officers and corporate directors are understandably wary of spending a lot of money to revamp a company's entire way of doing business. Even if they are willing to take this risk, they proceed very slowly to minimize possible losses. Corning reversed an erosion of its competitive position by gradually and cautiously implementing high-performance work systems. But the transformation took time. Companies with fewer resources and commitment are likely to shy away from the task.
To help lessen the risks for companies shifting to high-performance work systems and to speed the transition, federal agencies should incorporate best-practices demonstrations and test beds into their high-performance R&D programs whenever possible. Likewise, demonstrations and test beds in current government-sponsored programs and industry partnerships should be examined to determine the extent to which they can incorporate high-performance goals.
Diffusion. The diffusion of techniques and information is fundamental to overcoming companies' reluctance to change. One way to encourage new ideas is to establish standards and awards, such as the Malcolm Baldrige National Quality award. Government technology transfer and extension programs are also effective conduits and should be expanded and reoriented toward high-performance criteria. For example, NIST's Manufacturing Extension Partnership (MEP), through its Workforce Program, is helping small and medium-sized enterprises integrate workforce development and participation into their modernization efforts.
The government can also encourage the new trend toward high-performance industrial networks, whereby firms pool their resources to achieve economies of scale and scope, making it easier to overcome the barriers to high-performance work systems. MEP is already supporting some relevant workforce development activities through industrial network projects. Finally, it would be helpful to develop and disseminate curricula, textbooks, handbooks, and other educational materials on high-performance work systems, with a special emphasis on practices that can be applied to technology design, development, and implementation.
High-performance criteria and standards should be articulated and incorporated in national and international economic performance and quality standards, through mechanisms similar to the Baldrige award and related state quality awards, the National Medal of Technology, and production-standard certification systems such as ISO 9000 and the auto industry's QS 9000.
At the same time, MEP's workforce-development and labor-participation programs should be expanded. NIST also should provide merit-based awards for the development and deployment of technological tools, techniques, training curricula, and practices that improve the capacity of its centers to help small manufacturers. A small portion of economic development, labor training, and technology development funding should be set aside specifically for projects fostering and utilizing high-performance industrial networks.
Extension centers should also establish information services specifically devoted to helping businesses, labor unions, and public officials find resources helpful in implementing high-performance work systems. A High-Performance Technology Clearinghouse should be established as a center for information services and technology brokers. It could be set up at NIST or even contracted out to a commercial concern.
Meanwhile, NSF and the Departments of Commerce, Defense, and Education should institute programs that introduce education and training materials and curricula on high-performance work systems into the nation's engineering and business schools. The same can be done for community college technical courses; university labor education departments; labor union apprenticeship programs; and federal, state, and private sector job training programs.
Coordination. The above recommendations would infuse high-performance goals and criteria into the federal R&D effort. It is not a straightforward undertaking, however, given the diversity of government R&D activities and policies. It will take some care to ensure that the effort remains coordinated. Other policy initiatives that face the same challenge have been effectively coordinated with an interagency mechanism, such as the High-Performance Computing and Communications Program and initiatives on science and math education, global warming, and advanced materials. A national multiagency, multidisciplinary, High-Performance Technology Initiative should be created under White House auspices to provide coherence and coordination across all federal programs relevant to high-performance work systems. In addition, to ensure that high-performance goals and criteria are appropriately incorporated into government work, R&D agencies should broaden their advisory and review panels to include worker representatives.
Technology policy alone cannot ensure the widespread adoption of high-performance work systems. No government policy can. Ultimately, U.S. companies and workers will determine whether this new way of organizing work is beneficial. But government can be a potent partner in this crucial enterprise. For the sake of future U.S. jobs, economic growth, and competitiveness, each player-companies, workers, and government-must do its part to make the final outcome a success.
P. S. Adler and T. Winograd, eds., Usability: Turning Technologies into Tools. New York: Oxford University Press, 1992.
Commission on the Skills of the American Workforce, America's Choice: High Skills or Low Wages. Washington, D.C.: National Center on Education and the Economy, June 1990.
C. Ichniowski, T. Kochan, D. Levine, C. Olson, and G. Strauss, "What Works at Work: Overview and Assessment." Industrial Relations, Vol. 35, No. 3, July 1996.
E. Lawler III, The Ultimate Advantage: Creating the High-Involvement Organization. San Francisco, Ca.: Jossey-Bass, 1992.
S. Ralls, Integrating Technology with Workers in the New American Workplace. Washington, D.C.: U.S. Department of Labor, Office of the American Workplace, 1994.
H. Salzman and R. Lund, "Engineering Principles and Production Technology: Automation and Skill-Based Design in the U.S.," International Journal of Vehicle Design, Vol. 16, Nos. 4/5: 314-338, 1995.
B. Turner, M. Kaminski, D. Bertelli, M. Moye, and J. Yudken, Making Change Happen: Six Cases of Unions and Companies Transforming Their Workplaces. Washington, D.C.: Work and Technology Institute, 1996.
D. M. Upton, "What Really Makes Factories Flexible?" Harvard Business Review, July-August 1995: 74-84.
U.S. Department of Labor and U.S. Department of Commerce, Commission on the Future of Worker-Management Relations, Fact Finding Report. Washington, D.C.: U.S. Department of Labor, May 1994.
Kenan Patrick Jarboe and Joel Yudken are senior fellows at the Work and Technology Institute in Washington, D.C., and authors of the WTI report Smart Workers, Smart Machines: A Technology Policy for the 21st Century. Yudken is currently on leave from WTI, serving as senior adviser on Modernization and Workforce Development at the National Institute of Standards and Technology's Manufacturing Extension Partnership.