The measurement of output, prices, and productivity

Workers box jars of pasta sauce at a plant run by Chelten House Products in Bridgeport, New Jersey July 27, 2015.  The company produces private-label sauces, salsas and salad dressings for grocers such as Kroger, Whole Foods and Trader Joe's. As the Federal Reserve puzzles over what is holding back U.S. wages and productivity six years into the economic recovery, a pasta sauce company in New Jersey may offer some answers. Picture taken July 27, 2015. To match Insight USA-ECONOMY/LABOR REUTERS/Jonathan Spicer - GF20000007221
Editor's note:

This report is the product of the Productivity Measurement Initiative under The Hutchins Center on Fiscal and Monetary Policy at The Brookings Institution. The Intiative responds to the need and appetite for (a) examination and clarification of the concepts, purpose and relevance to policy debates of the “output” that comprises the numerator in the productivity measure, e.g. GDP vs welfare and (b) particular attention to the difficult measurement issues in rapidly changing, harder-to-measure and growing sectors of the economy, e.g. health care and information services.

In a dynamic economy, the nation’s statistical system faces a perennial challenge in accounting for the effects of innovation on new products and quality changes. Despite efforts to stay abreast of these changes, the view of many economists is that mismeasurement has led to long-term understatement of changes in living standards and productivity.

The most critical issues in productivity measurement relate to the measurement of price change. Price indexes are used to determine how much of the change in the sales of a particular good or service is due to price changes and how much is due to changes in real output, that is the quantity or quality of that good or service.

For this method to work, the price index needs to hold quality constant, but that can be very difficult to do. How much of the increase in spending on health care, for instance, reflects better quality as opposed to higher prices for the same services?  Unmeasured improvements in quality can have significant impacts on long-term trends in measured productivity

In 1996, the Advisory Commission to Study the Consumer Price Index, known as the Boskin Commission after its chairman, Michael Boskin of Stanford University, estimated that, relative to a true cost-of-living index, the CPI had an upward bias of 1.1 percentage points per year, with a plausible range of 0.8 to 1.6 percentage points.

In “The Measurement of Output, Prices, and Productivity: What’s Changed Since the Boskin Commission?” (PDF), Brent R. Moulton examines efforts made since 1996 by the US statistical agencies—especially the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA) to improve the price and volume statistics that are used in calculating real gross domestic product (GDP) and productivity measures.  He concludes that the overall bias of the CPI has fallen from about 1.1 percent to about 0.85 percent today.

The BLS, for instance, has overhauled the CPI, adopting a better method for estimating lower-level price indexes and improving sampling procedures, while expanding the use of hedonic methods for quality adjustment.  The BLS also has greatly expanded the Producer Price Index’s (PPI) coverage of services, allowing for improved analyses of industries and full industry coverage for multifactor productivity analysis. And the BEA has upgraded the GDP measure , with new coverage of investment in intangible intellectual property and improved measurement of hard-to-measure categories such as financial services. There are still steps, Moulton argues, that the agencies can take to improve the accuracy of their measures. Among them are:

  1. BLS and BEA should expand and reconstitute work on the digital economy, focusing particularly on improving the measurement of quality-adjusted prices for information and communications technology (ICT) equipment and associated digital services.
  2. BLS should reconsider its practice in dealing with product substitution or change by limiting most quality adjustments to cases when an item has disappeared from its sample.
  3. BLS and BEA should make a concerted, systematic effort to account for the effects of globalization in the measurement of GDP, value added by industry, and productivity.

Accurate statistics are critical for effective implementation of a wide range of economic policy—from monetary and fiscal policy for countercyclical macroeconomic policy, to providing cost-of-living adjustments for social security recipients, to providing accurate data for economic research.

Improving economic statistics must involve not only the work of the statistical agencies, he says, but also the engagement of the economics profession.

Read the full report here.

The author did not receive financial support from any firm or person with a financial or political interest in this article. He is not currently an officer, director, or board member of any organization with an interest in this article.