Why Efforts to Implement High-Involvement Work Practices FailPublished: October 13, 1999 in Knowledge@Wharton
Frits K. Pil, professor and research scientist at the University of Pittsburgh, and John Paul MacDuffie, professor of management at Wharton, explore the diffusion and use of best practice work systems in manufacturing in a new study. Despite clear evidence that high-involvement work practices and policies result in superior quality and productivity, the authors find tremendous variation in the use of such practices. Pil and MacDuffie explore why efforts to implement high-involvement work practices fail when the positive outcomes prove so compelling. The study appears in the book Employment Practices and Business Strategy edited by Peter Cappelli, director of the Center for Human Resources at Wharton.
The authors define "high-involvement" work systems as having combinations of five characteristics: On-line work teams, problem-solving groups, job rotation, suggestion programs, and decentralization of quality efforts. Research has shown that implementing any or all of these work practices can result in considerable improvement in overall performance, especially in labor-intensive settings such as manufacturing. Improved quality, higher productivity, lower scrap rates, and lower turnover are just some of the results seen when these work practices are used successfully.
Focusing on the global automobile industry, Pil and MacDuffie found that companies have been working assiduously to implement high-involvement work systems, but change is difficult to implement. The old way of doing things has to be unlearned, a new way learned, and then the new practices have to become routine before any positive results appear. Indeed, during the learning process, performance often drops to levels lower than those experienced prior to the change. While the long-term benefits of change may be clear, managers are often unwilling to suffer through short-term consequences. The authors found that, contrary to their expectations, a new management team was not the solution to successful change. Management teams with longer in-plant experience were more likely to introduce new work practices. In part this may be the result of a better understanding of operating methods, long-term working relationships with unions, and higher levels of employee trust.
Pil and MacDuffie found that three factors drive the success of efforts to adopt high-involvement practices:
- The level of complementary organizational practices and technologies that enhance the benefits from the new practices (e.g. flexible automation, such as robotics, enables companies to harness the workforce flexibility of high-involvement systems in ways that hard/fixed automation does not).
- The performance levels the organization is achieving from its current practices. Poor performers were more likely to undertake change than mediocre performers. The latter were more likely to focus on squeezing additional performance out of existing practices.
- Organizational actions that alter the cost of introducing new practices. As an example, the authors found that unfreezing of existing practices happens naturally at many companies (e.g. during new product introductions), and that many organizations use these occasions to undertake change. They urge caution, however, in using these opportunities, because factors that result in unfreezing of existing practice also make it difficult to identify performance changes from introducing new practices. Introducing new work systems in periods of stability makes it easier to focus attention on optimizing the practices, and assessing their effect on performance.
While the notion of universally applicable work systems is appealing, the authors use the Japanese automobile transplants in North America to show that adapting and tailoring practices to the local environment is not antithetical to achieving superior performance. The authors find, for example, that while the transplants make extensive use of team-work, team members undertake different tasks from those at plants in Japan. Similarly, while training levels are equally high, training topics differ. On the surface, the work systems at the transplants are similar to those used in Japan, but the actual implementation differs significantly.
By drawing on a world-wide sample of automobile production facilities, the authors show how government and union players can help facilitate (and, in some instances, hinder) the transition to high-involvement work systems. They also provide a brief discussion of the mechanisms companies use to transfer best practice internally between divisions and departments. Only a subset of these are effective, and few companies have been successful at implementing best-practice work systems throughout their organization.
Although the organizational benefits of shifting to high involvement work can be considerable, effective implementation often proves elusive. The professors stress that "Changing to high-involvement work is hard. Successful change is rare, and variance in work practice is widespread. Simply making the decision to implement suggestion boxes, job rotation, or work teams will not automatically lead to success." There is a complex interplay of plant-level, company-level, union-level, and country-level factors that influence the effective implementation of these work practices. But Pil and MacDuffie make clear that companies should engage all those involved in efforts to overcome the obstacles and implement high-involvement work systems. The long-term benefits invariably outweigh the costs.