Productivity Measures for Retail Trade: Data and Issues: Alternative Concepts for the Output of Retail Trade Industries Have Been Proposed and Implemented, but the Conclusion of Strong Labor Productivity Growth in This Sector Is Robust to the Choice of Measure

Manser, Marilyn E., Monthly Labor Review

The retail trade industry is a major component of the U.S. economy, with employment exceeding that in manufacturing. Yet only recently has the strong productivity performance of the retail sector been widely noted. Analysis of productivity growth in retail trade is especially challenging because it involves defining what output is for the industry, and different concepts can be used.

This article discusses conceptual and other issues in measuring productivity for retail trade industries, and presents current information on productivity in these industries in the United States. First it discusses the classification of retail trade activities. Second, it focuses on issues in defining the output concept for retail trade and in obtaining operational measures. Third, it presents data and comparisons for various measures and fourth, it addresses issues in comparing changes in retail trade productivity across countries.

Classification of retail trade

Recently, U.S. data have been converted to the new North American Industry Classification System (NAICS). NAICS retail trade (industries 44-45) includes stores and nonstore retailers and excludes food services and drinking places (NAICS 722). Both NAICS and the earlier Standard Industrial Classification (sic) system classify retail stores according to the types of goods that are being moved to the consumer. Eating and drinking places were classified as part of the retail trade division under the SiC (industries 52-59).

Two other major differences between NAICS and SIC also affect the classification of retail trade. Under NAICS, unlike SiC, auxiliary units involved in management or support activities such as transportation, warehousing, accounting and related services, and repair and maintenance are classified into specialized industries rather than including them in the industries they support, including retail trade. In addition, NAICS considers the method of selling when classifying establishments into wholesale versus retail trade, whereas the SIC system focused on the class of customer. This latter change caused a noticeable increase in the size of the retail trade sector, with a corresponding decrease in wholesale trade.

All of the changes previously described were introduced with the initial 1997 version of NAICS and continue under the 2002 NAICS revision. There were no changes at the three-digit level for retail trade between 1997 and 2002, but additional detail is provided for two retail trade industries under NAICS 2002. This article uses the NAICS 2002 definition of retail trade, unless otherwise noted.

Output concepts and issues

Industry output and productivity concepts. The broadest measure of productivity is multifactor productivity, which relates output to an index of all the inputs used in its production. Multifactor productivity (MFP) change measures the joint influences on economic growth of technological change, efficiency improvements, returns to scale, and other factors. The most commonly used measure of productivity is labor productivity, which is defined simply as output per hour. Labor productivity measures are produced and used more widely than multifactor measures because data needed for inputs other than labor are not available on a quarterly basis, are not as timely, and are not available or not measured precisely for many industries. Besides measuring the joint influences of the factors noted, labor productivity change also reflects the substitution of other inputs for labor in the production process.

To measure either labor or multifactor productivity for industries, we must define output. The Bureau of Labor Statistics (BLS) prefers the sectoral output concept for measuring industry output and productivity growth in the United States. Sectoral output is defined as gross output of the industry less intraindustry transfers. (1) This choice arises from the recommendation of the U.S. National Academy of Sciences that manufacturing industry multifactor productivity measures include intermediate inputs along with capital and labor inputs, as well as from subsequent research. An alternative output measure is value-added output, which is equal to sectoral output minus all intermediate inputs.

For retail trade, there are additional considerations. Another output measure, uniquely used for the distributive trades, is gross margins, which equals sectoral output less the cost of goods sold. It equals the sum of operating costs including other intermediate goods, other inputs, and residual profits.

In general, a retail trade industry produces sectoral output using labor hours, capital services, goods purchased for resale, and (other) intermediate inputs. The most general representation of the production process is:

f(O, M, I, K, L) = 0.

Often, we write:

O = g(M, I, K, L),

where:

O is the quantity of sectoral output;

M is the amount of goods purchased for resale;

I is the amount of intermediates;

K is the input of capital services

L is the input of labor hours

[p.sub.i] is the price of i, i = O, M, I, K, L.

Thus, gross-margin based output (GMO) is O - M, and value-added output (VA) is O - M - I. (2)

It is often noted in studies of production and productivity that use of a value-added measure assumes a type of separability between intermediate and other inputs, and does not allow for substitution possibilities between I, K, and L. In effect, beginning a study with gross-margin-based output does the same, allowing for no substitution among the types of goods purchased and other inputs. Ideally, one would like to have data on all the components of the model to test which is the appropriate approach. Unfortunately, sufficient information for these purposes is not available at present. In the absence of complete information on all the inputs identified, which is needed to produce a multifactor productivity measure, it is worthwhile considering merits and availability of alternative output measures for use in productivity studies.

There are a variety of conceptual issues affecting retail trade output, and various researchers have favored one or the other of the alternative concepts for output of this sector. Walter Oi argues that "The principal function of a retailer is to transfer possession of its merchandise lines to the ultimate consumer. (3) It assembles a product line, displays it at a convenient place and time, and provides ancillary services...." Generally, a retailer takes ownership of the products until they are sold to consumers. Thus, sales (sectoral output) is a commonly used measure of output for this industry sector. On balance, Oi prefers the sectoral output approach, rather than the gross margins approach, for measuring retail sector output for two primary reasons. (4) First, as already noted,"[t]he principal 'product' is a flow of transactions that transfers possession to material goods." Second, gross margins are the sum of value added (payments to labor and capital) plus monopoly rents. Changes in gross margins, like changes in sales, appropriately will reflect changes in efficiencies in store operations. But as Oi notes, gross margins also can increase as a result of businesses securing market power. (5) Increased monopolization, however, does not reflect increased output provided to consumers. (6) In contrast, …

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Productivity Measures for Retail Trade: Data and Issues: Alternative Concepts for the Output of Retail Trade Industries Have Been Proposed and Implemented, but the Conclusion of Strong Labor Productivity Growth in This Sector Is Robust to the Choice of Measure. Contributors: Manser, Marilyn E. - Author. Journal title: Monthly Labor Review. Volume: 128. Issue: 7 Publication date: July 2005. Page number: 30+. © 1999 U.S. Bureau of Labor Statistics. COPYRIGHT 2005 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.