Sale signs are displayed on racks of clothing outside a Rainbow clothing store along the Fulton Street Mall in the Brooklyn borough of New York, U.S. (Photographer: Victor J. Blue/Bloomberg)

U.S. Stats Officials Say Measurements of GDP, Inflation Are Off

(Bloomberg) -- Top officials from two U.S. government economic-statistics agencies said their measurement tools are understating growth and overstating some components of inflation by modest amounts, while cautioning that this doesn’t explain the sluggish expansion in recent years.

“The Bureau of Labor Statistics and Bureau of Economic Analysis agree that price index mismeasurement continues to lead to understated growth in real output over time,” five current and former officials from the agencies wrote in a paper published May 3 in the American Economic Association’s Journal of Economic Perspectives and presented last week at a meeting of the BEA’s advisory committee.

Economists for years have questioned whether statistical agencies are keeping up with changes, resulting from an increasingly digitized and innovation-based economy, that improve the quality of goods and services. The answer: It’s hard, especially in technology and medical services where innovations can be rapid and the results hard to capture. Health-care spending represented about 17.5 percent of gross domestic product in 2014, they note.

Despite the BLS’s improvements in capturing changes in the economy, “our quality corrections are not as complete as they could be, so they are not taking into account some of the innovations that are actually going into the goods,” said Erica Groshen, one of the paper’s authors, who served a four-year term as BLS commissioner that ended in January.

The methods for improved accuracy haven’t been implemented for reasons including “high computational intensity and cost” and the need for larger sample sizes, according to the paper.

The U.S. economy has grown by about 2.1 percent a year during the expansion that began in 2009, compared with 2.8 percent during the previous expansion, according to the BEA.

Groshen’s co-authors include Brian Moyer, director of the Bureau of Economic Analysis, which publishes the GDP data; BEA senior economist Ana Aizcorbe; and two BLS officials, Ralph Bradley and David Friedman. The BLS publishes data including consumer prices and employment.

Price Measures

Looking at the personal consumption component of GDP, the authors find that price measures likely show an overstatement of 0.2 percentage point in 2000, rising to 0.26 percentage point in 2015. If inflation is overstated, real GDP is understated.

Combined with some mismeasurement of quality changes in computer and software investment, the total effects would mean GDP growth is understated by about 0.4 percentage point for 2000, 2005, 2010 and 2015, according to the paper.

The authors say that measurement issues are “far from new” and the magnitude and timing of their findings are unlikely to account “for the pattern of slower growth in recent years.”

Still, the overstatement of inflation would raise an important policy question for the Federal Reserve, which U.S. central bankers themselves are asking: Is the 2 percent target too low?

On an annual basis, the personal consumption expenditures price index has risen 2 percent or more just once since April 2012, which is one reason why Fed Chair Janet Yellen has raised interest rates just three times since December 2015. The economic recovery began in June 2009.

Inflation Picture

“For GDP and productivity growth,” the paper’s conclusions “make a lot of sense but don’t really change the overall picture of a productivity slowdown,” said Julia Coronado, president of MacroPolicy Perspectives LLC in New York and a former Fed economist. “It is more of a game changer for thinking about inflationary pressures.”

Groshen, a former New York Fed economist, said it’s hard to conclude what the optimal rate of inflation should be. Economists generally agree that it should be low, and consumers and businesses should have confidence that it will remain so.

“If there are bigger measurement issues than people had appreciated, you could argue for a higher target, and a bigger band of tolerance around it,” she said in an interview Monday.