How do they measure productivity?

How do they measure productivity?

How do they measure productivity?

Answers to your questions about the news.
Dec. 3 2003 4:12 PM

How Productive Are You?

The ways the government keeps tabs on American workers.

Illustration by Mark Alan Stamaty

The latest Department of Labor statistics show that productivity rose at a 9.4 percent annual rate last quarter, the quickest pace in 20 years. What does the productivity figure mean, exactly, and how is it calculated?

Productivity is meant to show how efficiently the American labor force is churning out goods and services, as indicated by the nation's economic output per hour worked. The first part of the equation, output, is similar to gross domestic product, though not identical. The Bureau of Labor Statistics looks at myriad industries and estimates the monetary value of what each one produces. Output for the automobile industry, for example, is gauged according to the value of all cars manufactured, while newspaper output stems from combined advertising and subscription revenues. Unlike GDP, however, which tots up the total dollar value of all goods and services, the BLS output estimates exclude industries for which output is less tangible, such as the government, nonprofits, and domestic help (e.g., nannies and maids).


The BLS thus comes up with a few different sets of total output figures: business output, nonfarm business output, and manufacturing output. The next step is to divide those figures by the "input," or the number of hours worked. This data primarily comes from the bureau's monthly survey of business owners, part of the Current Employment Statistics program. These surveys only cover nonagricultural work, however; to measure hours worked on farms, the BLS instead turns to data from the monthly Current Population Survey, conducted in conjunction with the Census Bureau.

The biggest guessing game involves the hours worked by supervisors, whose contributions aren't noted in the CES survey. In nonmanufacturing sectors (like the newspaper biz), supervisors are assumed to work the same amount of hours as their charges; in manufacturing, the rule of thumb for white collars is 40 hours a week. Again, the BLS comes up with hours worked for total business, nonfarm business, and manufacturing, and then divides the output by the hours worked. The 9.4 percent figure quoted in today's news reports applies only to nonfarm business; the total business number is a still robust 8.6 percent. The sharp uptick is largely due to gains in output, rather than hours worked.

Explainer thanks Katharine G. Abraham of the University of Maryland.