, 1:663-698. New York: The Oxford University Press, 2018.
The World KLEMS Initiative was established at the First World KLEMS Conference at Harvard University in August 2010[i]. The purpose of this Initiative is to generate industry-level data on outputs, inputs, and productivity. Productivity is defined as output per unit of all inputs. The inputs consist of capital (K) and labor (L), the primary factors of production, and intermediate inputs of energy (E), materials (M), and services (S). The acronym KLEMS describes these inputs. Industry-level data have been proved to be indispensable for analyzing the sources of economic growth for countries around the world.
International productivity comparisons are the second focus of industry-level productivity research. Productivity gaps between two countries are defined in terms of differences in productivity levels. These differences are measured by linking the productivity levels for each country by purchasing power parities for inputs and outputs. As an example, the purchasing power parity for Japan and the U.S. is defined as the price in Japan, expressed in yen, relative to the price in the U.S., expressed in dollars. Purchasing power parities can be defined in this way for commodities, industries, or aggregates like the GDP. Productivity gaps are essential for assessing competitive advantage and designing strategies for economic growth.
We review productivity measurement at the industry level in Section 2. The landmark EU (European Union) KLEMS study was initiated in 20003 and completed in 2008. This study provided industry-level data sets for the countries of the European Union. These data have proved to be invaluable for analyzing the slowdown in European economic growth. The EU KLEMS study also included data for Australia, Canada, Japan, Korea, and the United States. These data have been widely used for international comparisons between European countries and the leading industrialized countries of Asia and North America.
Regional organizations – LA KLEMS in Latin America and Asia KLEMS in Asia – have joined the European Union in supporting industry-level research on productivity. The Latin American affiliate of the World KLEMS Initiative, LA KLEMS, was established in 2009 at the Economic Commission for Latin American and the Caribbean (ECLAC) in Santiago, Chile. The Asian affiliate, Asia KLEMS, was founded at the Asian Development Bank Institute (ADBI) in Tokyo in 2010. The regional organizations have stimulated the development of industry-level productivity measures for the emerging economies of Asia and Latin America, such as Brazil, China, and India, as well as measures for the advanced economies of Asia, Europe, and North America.
In Section 3 we present the KLEMS framework for productivity measurement for a single country. Development of this framework within the national accounts has the important advantage that official measures can be generated at regular intervals in a standardized format.[ii] The production account in current prices contains nominal outputs and incomes, while the production account in constant prices provides real outputs and inputs, as well as productivity. Paul Schreyer’s (2001) OECD Productivity Manual provided methods for productivity measurement within the national accounts.
A key feature of the KLEMS framework is a constant quality index of labor input that combines hours worked for different types of labor inputs by using labor compensation per hour as weights. Similarly, a constant quality index of capital input deals with the heterogeneity among capital services by using rental prices of these services as weights. Schreyer’s (2009) OECD Manual, Measuring Capital, presented methods for measuring capital services. Finally, inputs of energy, materials and services are generated from a time series of input-output tables in current and constant prices.
In 2008 the Advisory Committee on Measuring Innovation in the 21st Century to the U.S. Secretary of Commerce recommended that productivity data be incorporated into the U.S. national accounts. This was successfully completed by the Bureau of Economic Analysis (BEA), the agency responsible for the U.S. national accounts, and the Bureau of Labor Statistics (BLS), the agency that produces industry-level measures of productivity for the U.S. Susan Fleck, Steven Rosenthal, Matthew Russell, Erich Strassner, and Lisa Usher (2014) published an integrated BEA/BLS industry-level production account for the U.S. for 1998-2009 in Jorgenson, Landefeld, and Schreyer (2014).
In Section 4 we illustrate the KLEMS methodology for a single country by summarizing the industry-level productivity data for the United States for the period 1947-2012 compiled by Jorgenson, Ho, and Samuels (2016). We analyze the sources of U.S. economic growth for three broad periods: the Postwar Recovery of 1947-1973, the Big Slump of 1973-1995, following the energy crisis of 1973, and the period of Growth and Recession, 1995-2012. To provide more detail on the period of Growth and Recession, we analyze the sources of growth for the sub-periods 1995-2000, 2000-2007, and 2007-2012 – the Investment Boom, the Jobless Recovery, and the Great Recession.
In Section 5 we introduce the KLEMS framework for international comparisons by presenting price level indices and productivity gaps. The price level index is an indicator of international competitiveness, often expressed as over- or undervaluation of currencies. A specific example is the over- or undervaluation of the Japanese yen relative to the U.S. dollar.
The price level index for Japan and the United States compares market exchange rates with purchasing power parities for the GDP.
The productivity gaps between Japan and the U.S. are indicators of the relative efficiency of two countries in transforming inputs into outputs. To measure these productivity gaps we first construct comparable measures of productivity. We then link the U.S. and Japanese outputs and inputs at the industry level by means of purchasing power parities. As an illustration, the U.S. productivity data presented in Section 4 for 1947-2012 have been linked to comparable Japanese productivity data for 1955-2012 by Jorgenson, Nomura, and Samuels (2016).
The international comparisons between Japan and the U.S. presented in Section 6 are based on industry-level purchasing power parities. These comparisons provide important information on the valuation of the Japanese yen relative to the U.S. dollar. The yen was under-valued from 1955 until the Plaza Accord of 1985. This enabled Japan to achieve a high level of international competitiveness, despite a large productivity gap with the United States. Since 1985 the yen has been over-valued, relative to the dollar, reaching a peak in 1995 that greatly undermined Japanese competitiveness. The yen finally achieved purchasing power parity with the dollar only in 2015, restoring Japanese international competitiveness after several years of monetary policies based on quantitative easing by the Bank of Japan.
The large productivity gap between Japan and the United States that existed in 1955 gradually closed until the end of the “bubble economy” in Japanese real estate in 1991. Since that time Japanese productivity has been stagnant, while productivity in the U.S. has continued to rise. The widening productivity gap can be traced to a relatively small number of industrial sectors in Japan, mainly in trade and services, but also including agriculture. Productivity gaps for Japanese manufacturing industries have remained relatively small. This has created opportunities for formulating a Japanese growth strategy based on stimulating productivity growth in the lagging industrial sectors. Section 7 presents our conclusions.