We thank Bernhard Pachl, Ramona Schrepler (who did all of the programming) and Lars Siemers for excellent research assistance, and Alexandra Holten for the skilful and patient typing of a trying manuscript.We are also grateful to participants in several seminars at the World Bank for their valuable comments and suggestions, while absolving them of responsibility for any errors of opinion or analysis that remain. Finally, the opinions expressed in the paper are ours as individual scholars, and not necessarily those of our respective institutions.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. This content downloaded from 146.ON AVERAGE, manufacturing productivity (measured by real value added per hour) is higher in U.S. operations than in Japanese or European operations, but productivity by industry varies substantially from country to country.' What is not well established is why such international productivity differences exist. This paper explores international productivity differences in manufacturing industries across Germany, Japan, and the United States and offers an explanation for them.We develop an extension and a refinement of the industry-of-origin method to provide new measures of productivity in nine industries located in the three countries (automobiles, automotive parts, metalworking, steel, computers, consumer electronics, food, beer, and soap and detergent). Then we combine publicly available data with the industry knowledge and assessments by experts within McKinsey & Company to determine the reasons for productivity differences. The nature of the explanation takes place at two levels, the first of which is at the production process level. We look for differences in the use of capital, technology, and skills-how the variables that enter the production As an illustration, consider the following two examples. First, in the 1980s German industry, on average, fell behind best-practice methods despite the competition it faced within Europe. During the 1960s and 1970s, manufacturers in Germany raised productivity rapidly, achieving almost complete convergence with the United States in manufacturing productivity. But in the 1980s, German industries still concentrated on competition in Europe, where they had a significant productivity advantage, and did not compete against the emerging productivity leaders in Japan in the same way as U.S. companies did. For example, direct competition from leading Japanese producers caused the exposed U.S. automobile industry to respond with high productivity, while by 1990 the German industry-much more sheltered from Japanese competition-had done little to increase its productivity. Second, we observed marked productivity disparities within Japan. Those parts of Japanese manufacturing that competed in world markets had achieved best-practice methods or close to it. Those industries that competed domestically had relatively low productivity, even though the domestic industry could be quite competitive.This study parallels the findings of the earlier McKinsey study on productivity in the service sector and adds to the validity of those results by finding a similar effect of competition on productivity in the manufacturing sector.2 Moreover, we were able to be more specific and detailed in describing how productivity differenc...

No abstract

We examine whether the publication of the individual voting records of central-bank council members is socially beneficial when the public is unsure about the efficiency of central bankers and central bankers are angling for re-appointment. We show that publication is initially harmful since it creates a conflict between socially desirable and individually optimal behavior for somewhat less efficient central bankers. However, after re-appointment, losses will be lower when voting records are published since the government can distinguish highly efficient from less efficient central bankers more easily and can make central bankers individually accountable. In our model, the negative effects of voting transparency dominate, and expected overall losses are always larger when voting records are published.