Technical Notes

The Bureau of Labor Statistics constructs trend indexes of manufacturing labor productivity (output per hour), hourly compensation costs, and unit labor costs from three basic aggregate measures-output, total labor hours, and total compensation. The hours and compensation measures refer to all employed persons, including self- employed persons and unpaid family workers, in the United States, Canada, Japan, France, Germany, Norway, and Sweden and to all employees (wage and salary earners) in the other economies. For all of the countries, the term "hours" refers to hours worked.

In general, the measures relate to total manufacturing as defined by the International Standard Industrial Classification (ISIC). However, the measures for France and Italy (beginning in 1970) refer to mining and manufacturing less energy- related products, and the measures for Denmark include mining and exclude manufacturing handicrafts from 1960 to 1966.

Output. For most countries, the output measures are value added in manufacturing from the national accounts. However, output for Japan prior to 1970 and for the Netherlands prior to 1960 are indexes of industrial production. The national accounts measures for the United Kingdom are essentially identical to their indexes of industrial production. While methods of deriving national accounts measures differ from country to country, BLS has reviewed these methods and determined that the series are sufficiently comparable for measuring comparative trends in productivity and unit labor costs.

In this release, the 1977-97 manufacturing output data for the United States are the gross product originating (value added) measures prepared by the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce. Comparable manufacturing output data currently are not available prior to 1977.

U. S. gross product originating is a chain-type annual-weighted series.(1) The Japanese value-added series is based upon one set of fixed price weights for the years 1970 through 1997. Output series for most other foreign economies also employ fixed price weights, but the weights are updated periodically (for example, every 5 or 10 years).

To preserve the comparability of the U.S. measures with those of the other economies covered, BLS uses gross product originating in manufacturing for the United States for these comparative measures. The gross product originating series differs from the manufacturing output series that BLS publishes in its news releases on quarterly measures of U.S. productivity and costs. The quarterly measures are on a "sectoral output" basis rather than a value-added basis. Sectoral output is gross output less intra-sector transactions.

Labor Input. The total hours measures are developed from statistics of manufacturing employment and average hours. The series used for France (from 1970 forward), Norway, and Sweden are official series published with the national accounts. Where official total hours series are not available, the measures are developed by BLS using employment figures published with the national accounts, or other comprehensive employment series, and estimates of annual hours worked. For Germany, BLS uses estimates of average hours worked, developed by a research institute connected to the Ministry of Labor for use with the national accounts employment figures. For the other countries, BLS constructs its own estimates of average hours.

- 14 -

Denmark has not published estimates of average hours for 1994-1997 and, therefore, the BLS measure of labor input for Denmark ends in 1993. Adequate employment and average hours data are not available for Korea or Taiwan; therefore BLS does not calculate labor input for these economies.

Compensation (labor cost). The compensation measures are from the national accounts except those for Belgium, which are developed by BLS using statistics on employment, average hours, and hourly compensation. Compensation includes employer expenditures for legally required insurance programs and contractual and private benefit plans, in addition to all payments made in cash or in-kind directly to employees. For France and Sweden, compensation is increased to account for other significant taxes on payroll or employment. For the United Kingdom, compensation is reduced between 1967 and 1991 to account for subsidies. Self-employed workers are included in the all-employed-persons measures by assuming that their hourly compensation is equal to the average for wage and salary employees.

Current Indicators. The measures for recent years may be based on current indicators of output (such as industrial production indexes), employment, average hours, and hourly compensation until national accounts and other statistics, normally used for the long-term measures, become available.

Trade-Weighted Measures. The trade weights for Canada, Japan, and the European countries were obtained by rescaling a series of weights, developed by the International Monetary Fund, based on average trade flows over the 1989-91 period. These weights are based on aggregate trade data for total manufacturing and take account of both bilateral trade and the relative importance of "third country" markets. The 1989-91 weights do not include Korea and Taiwan. BLS developed weights for Korea and Taiwan by using data from an earlier study from the International Monetary Fund and other sources.

The following weights were used for the entire period for which
trade-weighted unit labor cost measures are produced:

Country      Weight      Country            Weight    

Canada        24.46      Italy                4.44
Japan         29.54      Norway               0.46
Belgium        2.06      Sweden               1.83
Denmark        0.46      United Kingdom       8.68
France         5.70      Korea                5.55
Germany       11.22      Taiwan               5.60

Level comparisons. The BLS measures are limited to trend comparisons. BLS does not prepare level comparisons of manufacturing productivity and unit labor costs because of data limitations and technical problems in comparing the levels of manufacturing output among countries. Each country measures manufacturing output in its own currency units. To compare outputs among countries, a common unit of measure is needed. - 15 -

Market exchange rates are not suitable as a basis for comparing output levels. What are needed are purchasing power parities, which are the number of foreign currency units required to buy goods and services equivalent to what can be bought with one unit of U.S. currency.

Purchasing power parities are available for total gross domestic product (GDP) from the Organisation for Economic Cooperation and Development (OECD). However, these parities are derived for expenditures made by consumers, business, and government for goods and services-not for value added by industry. Therefore, they do not provide purchasing power parities by industry. The parities for total GDP are not suitable for each component industry, such as manufacturing.

Some analysts have constructed "proxy purchasing power parities" for manufacturing, using selected expenditure items from the GDP purchasing power parities. However, the prices used in this method have a number of shortcomings because they are based on consumer, business, and government final expenditures. The data reflect sales to final users, rather than the output of specific industries, and thus prices include indirect taxes, distribution margins, and transportation costs. Further, the purchasing power parities exclude items which, although they are outputs of the manufacturing sector (e.g., steel, paper, cement, etc.), are only intermediate inputs with respect to final sales. They also do not reflect an industry's output because they include imports and exclude exports.

A different approach has been used by researchers at the International Comparisons of Output and Productivity project at the University of Groningen in the Netherlands. They develop unit value ratios for matched products, based on censuses of manufactures data in the various countries. Unit values are obtained by dividing producers' sales values by quantities shipped. This procedure also has shortcomings, primarily because only a limited portion of manufactured products can be matched and the matched products may not be adequately representative of total manufacturing. For many products, values are reported but not quantities; for some products, there is no counterpart in the other country; and other products cannot be matched because they represent a different mix of product varieties or because of large differences in qualities.(2)

_______________________

(1) For more information on the U. S. measure, see Robert E. Yuskavage, "Improved Estimates of Gross Product by Industry, 1959-94," Survey of Current Business, August 1996, pp. 133-55.

(2) For an article based on this approach, see Bart van Ark, "Manufacturing prices, productivity, and labor costs in five economies," Monthly Labor Review, July 1995, pp. 56-72.

Table of Contents

                              Back to Foreign Labor Statistics