(Bloomberg Businessweek) -- The consumer price index is one of the U.S. government’s most important instruments. The century-old gauge is designed to measure inflation, but the CPI is also used to calculate the economy’s “real” growth rate (that is, output adjusted for inflation). Social Security payments and federal government pensions are also pegged to it.
New research shows that the CPI is slow to reflect changes in prices—and, equally important, understates the degree to which prices move up and down. The problem stems from the way the government calculates the price of shelter, a category that makes up one-third of the index.
Three economists have developed an alternative measure that captures price moves as soon as they occur and shows the full range of changes. If it had existed in 2008-09, when the economy was in the deepest recession since the Great Depression, it would have shown far deeper deflation than the Bureau of Labor Statistics registered. The official CPI, they write in a new paper, was overstating inflation by 1.7 percentage points to 4.2 percentage points annually during the Great Recession. More recently, they write, the problem has been the opposite: Annual readings have understated inflation by 0.3 to 0.9 percentage points. Those are huge disparities given that forecasters make a big deal of fluctuations of just one or two tenths of a percentage point in the official rate.
The new yardstick is the latest in a long line of attempts to measure inflation better, among them the Billion Prices Project at MIT and Harvard and Adobe Systems Inc.’s Digital Price Index. The economists behind it are Brent Ambrose and Jiro Yoshida of Pennsylvania State University’s Smeal College of Business and Edward Coulson of the Merage School of Business at the University of California at Irvine. Their latest version is described in an April 20 academic paper titled “Housing Rents and Inflation Rates.” The key difference from the CPI is that their measure factors in only new rental leases, including those of new tenants and old ones who recently renewed. The BLS, in contrast, also includes rent paid by tenants whose leases weren’t up for renewal in the latest month, which means it’s slower to pick up on changes in market conditions.
The reason rents matter so much to the CPI is that the government doesn’t take into account home sales prices when calculating shelter costs. Instead, it figures how much the owner of a house would have to pay each month to live there—what it calls the “owner’s equivalent rent”—by looking at rental rates for houses of similar size, location, and quality. (That can be tricky for expensive houses, for which rental equivalents are hard to find, but that’s another story.)
In response to a query from Bloomberg Businessweek about the CPI, BLS spokesman Steve Reed issued a statement saying “we believe that our current methodology is appropriate, but we are open to reconsidering it based on new evidence.”
At the core of the discrepancy is a disagreement over the CPI’s proper function. Should it be a measure of what the overall public is paying for goods and services, or rather a timely indication of the amount of price pressure in the economy? The BLS measure is more the first, while the academics’ measure is more the second.
“The CPI is used for many different things,” Randal Verbrugge, an economist who’s doing consulting work for the BLS on inflation measurement, wrote in an email. For rate-setters at the Federal Reserve, the optimal CPI might be one that detects incipient shelter inflation, he suggests, while for the Social Security Administration, it might be better to have a CPI that’s a broader snapshot of shelter costs. (The Federal Reserve’s preferred measure of inflation is the personal consumption expenditures price index. That index gives less weight to shelter—about 16 percent, vs. about 33 percent in the CPI.)
CoreLogic, the Irvine, Calif.-based company that collects and sells information on properties, has created a monthly Single-Family Rent Index that combines the methodology used by the Penn State/UC Irvine team with data from real estate portal MLS.com. It foreshadows by exactly one year the shelter component of the consumer price index, according to Sam Khater, who until recently was CoreLogic’s deputy chief economist and now is chief economist at Freddie Mac. In an email before leaving CoreLogic in April, Khater wrote that based on the sneak peek provided by the Single-Family Rent Index, “there’s no reason to expect” an uptick in the official measure of shelter inflation in the coming year.
Verbrugge, the consultant, wrote in an email that while he had doubts about the MLS data he was still impressed by the predictive power of CoreLogic’s index. He said that he had no knowledge of any BLS plans to redesign the CPI, adding that it would be a costly endeavor at a time when the bureau is already facing budget pressures.
It’s hard to know how the Fed might have behaved if it had been using an index like the Penn State/UC Irvine one during the Great Recession. Deflation would have looked more severe, but the decline in gross domestic product would have looked less severe. The current expansion looks less healthy using the alternative measure because inflation comes out higher and growth lower. The combination “may explain the consistent negative consumer sentiment and political uncertainty seen during the economic expansion following the Great Recession,” the authors write.
To contact the author of this story: Peter Coy in New York at pcoy3@bloomberg.net.
To contact the editor responsible for this story: Cristina Lindblad at mlindblad1@bloomberg.net.
©2018 Bloomberg L.P.
.jpg?w=600)
