Retail Transformation Through Experimental Measures
On September 25, 2023, the Bureau of Labor Statistics (BLS) introduced experimental measures for Retail Trade Industries in response to the recommendations of the 2021 report on A Satellite Account to Measure the Retail Transformation from the Committee on National Statistics (CNStat). BLS commissioned CNStat to convene a panel of experts to get input on how better to understand the changes in the retail trade sector. Their report includes suggestions to develop retail margin productivity, pursue more timely data releases, and include activity from supporting industries. The content that follows is a first step in addressing these recommendations. BLS is exploring these alternative measures of output and labor productivity for retail industries to expand the understanding of our current annual data as featured in the Productivity and Costs by Industry – Wholesale Trade and Retail Trade Industries news release and Wholesale and Retail Trade Industries Labor Productivity highlights.
BLS welcomes feedback and suggestions for these experimental retail measures. Feel free to contact us to share your thoughts: https://www.bls.gov/productivity/contact.htm
1) Sectoral versus Gross Margin: Alternative Measures of Output and Labor Productivity for Retail Industries
Current annual BLS labor productivity data for retail industries use a real sectoral output approach to measure changes in output per hour worked, or labor productivity. Real sectoral output is measured as total sales revenue deflated by the prices of products sold outside the industry. Retail sales prices reflect the entire value chain of the product sold from product design to delivery to the final consumer.
Retail output can be considered as a service rather than the quantity of products it delivers to consumers. Gross margin output is measured as the difference of total sales less the cost of goods sold and reflects the services provided by retail establishments that help facilitate the sale of goods to the public. Margin prices, calculated as the difference between the sales price and the acquisition price, reflect the value added by retailers for services such as the procurement, storage, marketing, and display of goods for sale. Gross margin output isolates the services that retailers provide to consumers by excluding the quantity of goods flowing through retail establishments.
This page compares trends in real sectoral output and estimates of real gross margins and their corresponding labor productivity measures for three retail industries that touch our day-to-day lives: gasoline stations, grocery stores, and clothing stores.
2) Higher-frequency Output: Introducing the 12-Month Trailing Average
BLS publishes measures of real sectoral output, sectoral output, and sectoral output price deflator for retail trade industries in its Productivity and Costs by Industry Wholesale Trade and Retail Trade Industries Economic News Release. However, these values are currently only released on an annual basis. To better meet the needs of its customers, the BLS productivity program is exploring new higher-frequency real output measures.
Annual output values for retail industries are calculated using sales data from the Census Bureau’s Annual and Monthly Retail Trade Surveys (ARTS and MRTS) for all published years. Industry sales are then broken out by product line using ratios calculated from the Economic Census Product Line Sales, deflated by the corresponding Producer Price Index (PPI) or Consumer Price Index (CPI), and aggregated back to the industry level.
For this experimental measure, BLS continues to use Census, PPI, and CPI as the sources of its sales and deflator data but makes use of their monthly data instead of annual data. Although our annual measures utilize some custom-weighted deflators to better fit certain product line mixes, the experimental measure relies on directly published CPI and PPI data series. The same methods as the published measure are applied to the monthly data to create a monthly output index for each retail industry. These monthly indexes are volatile and reflect significant seasonality, which complicates analysis and comparison to BLS annual values. To reduce the effects of seasonality and provide a measure that is more comparable to our annual measures, BLS introduces an experimental 12-month trailing average output measure for retail trade industries.
The 12-month trailing average takes an average of an industry’s monthly real output value over the previous 12 months. For example, to calculate the 12-month trailing average real output value for January 2018, we take an average of the monthly real output values from February 2017 through January 2018.
This page compares output trends for three retail industries that demonstrate the added analytical and predictive power of the experimental measure: automobile dealers, grocery stores, and clothing stores.
3) Expanded Retail Sector
BLS seeks to better understand the implications for employment and productivity measurement in retail trade and to capture gains in productivity associated with shifts across industries. BLS has created experimental measures for an expanded retail trade sector by assigning shares of total work hours in transportation and warehousing (NAICS 48,49) to the retail trade sector (NAICS 44,45). These measures expand beyond the bounds of the North American Industry Classification System (NAICS) framework to include retail-supporting industries, such as transportation and warehousing.
The retail trade sector employed approximately 16.2 million people, or 12.5 percent of all total nonfarm employees in 2022, with 24.7 billion hours worked. In the same year, the transportation and warehousing sector employed approximately 8.0 million people, or 6.2 percent of nonfarm employees, with 15.2 billion hours worked.
Without any concrete data on how many workers in the transportation sector support retail trade specifically, we explore a range of assumptions starting from ten percent of transportation and warehousing hours worked up to 100 percent. The share values BLS allocated from the retail-supporting industries of transportation and warehousing may be construed as upper and lower bound concepts. The objective is to compare published measures of BLS hours worked and labor productivity series in the retail sector to these experimental series.
The charts below feature the progression of adding larger proportions of hours worked from transportation and warehousing to the retail sector.