Economist: The Fed Would've Hiked Rates Already If It Went Back to Its Old Measure of Inflation

One measure could make a big difference.

Janet Yellen, who heads the U.S. Federal Reserve.

Andrew Harrer/Bloomberg

The Federal Reserve doesn't have a deflation problem: It has a measurement problem.

So says Joe Lavorgna, chief U.S. economist at Deutsche Bank, who reckons that the Fed should go back to the days of using the Consumer Price Index rather than Personal Consumption Expenditures, as it did prior to the dawn of the new millennium. 

In 2000, former Fed Chairman Alan Greenspan opted to begin using PCE, arguing that PCE had a formula that better reflected the changing composition of spending, was more comprehensive, and would be more consistent over time by taking revisions into account. At the same time, he affirmed the central bank's commitment to monitoring a variety of gauges of price pressures, although the PCE index continues to reign supreme in the pantheon.

Lavorgna, however, contends that some of the features cited by Greenspan are actually bugs.

Here are six reasons he thinks CPI is the better metric. 

  1. Weighting Scheme. While CPI puts a large weight on housing costs, PCE puts a bigger emphasis on medical care. Lavorgna argues that for the majority of Americans, housing is the greater expense. 
  2. The PCE is an implicit, chain-price index, while the CPI is a fixed-weight index. In the former, weights change constantly to reflect spending shares, while the latter sees prices change primarily after the end of a given reference period. "An implicit price indicator will almost always run below a fixed price indicator, because if a household had a choice between substitutes that are similarly priced, they'll always tend to buy more of the lower priced good," explains LaVorgna.
  3. PCE gets revised and CPI does not.  The argument here is that you could see a large upward revision, leaving the Fed behind the curve when it comes to raising rates and inflation.
  4. Wage/cost of living adjustments all set to CPI or some variation thereof, not PCE. As such, the PCE is an index that isn't used in the marketplace and has less practical value. St. Louis Fed President James Bullard even noted this discrepancy in 2013, when he pointed out that the U.S. federal government uses CPI to make inflation adjustments regarding such benefits as Social Security and suggested that it would be beneficial for the government and the Fed to settle on a standard measure of inflation. 
  5. Inflation expectations inform monetary policy decisions. The "committee has taken note of recent declines in market-based measures of inflation compensation and will continue to monitor inflation developments carefully," said Janet Yellen, the current Fed chair, in her latest press conference. LaVorgna notes that a rich history surrounds inflation expectations and that forecasts such as the Philadelphia Fed's Survey of Professional Forecasters and the Blue Chip Economic Indicators Survey measure expected price pressures in terms of the CPI. Inflation-indexed bonds, in addition, are also based on the CPI.
  6. Imputation and miscellaneous idiosyncrasies. The PCE index has some other undesirable quirks, the economist notes. Chief among them is the increased use of imputed rather than directly observed data—for instance, in the case of financial service costs. The PCE index also monitors changes in prices of goods and services purchased by nonprofits, which LaVorgna contends is not a relevant input when tracking a household's cost of living.

This isn't a new argument, but it has gained steam as the PCE has continued to lag even further behind the Fed's 2 percent target than CPI, and the difference, or spread, between the two metrics widening.

In fact, the core index of the PCE is now lagging CPI by the most since the recession ended in June 2009.


Even though it appeared that global developments in markets overseas—such as China and other emerging economies—were a large reason the Fed decided to hold interest rates at historic lows in its latest policy decision, inflation is still one of the main focuses as "The Committee continues to monitor inflation developments closely," according to the Fed's most recent statement.  

If the Fed were still using core CPI as its primary operational guide for inflation, LaVorgna believes the near-zero interest rate policy in the U.S. would already be a thing of the past.

"We would have higher rates for sure. We might even have a healthier economy because I do believe that keeping rates low for as long as they have, there's a message in here, and that's hurt investor and business risk taking, not investor risk-taking as much as business risk-taking," he said in an interview. "Short rates would definitely be higher. Now, would the economy be better? We'll never know that."

The economist, however, doesn't think the Fed is poised to switch its preferred measure of inflation soon.

"Unfortunately, [PCE] is what we'll continue to have because the Fed can't say it's meeting its inflation mandate by changing the definition of inflation, which would be quite apparent if they were to do so," he said. "It's been undershooting its preferred inflation target for over three years, it's really hard to say, 'okay, well now this is the better series, and this is now close to two so we hit our inflation mandate.'"

LaVorgna foresees the gap between the two measures of inflation staying the same or even widening over the next couple of quarters, which raises the risk of a policy error by the central bank. 

The economist outlined a potential scenario that could arise in the near future, in which core CPI is comfortably above 2 percent, with low vacancy rates putting upward pressure on rents, while the PCE index continues to run at a sub-1.5 percent rate.

Core CPI "certainly would suggest the Fed should be raising rates, and possibly quite aggressively," he said. But still-subdued core PCE would mean "the Fed's not moving, even though I would argue in theory, it should."

Before it's here, it's on the Bloomberg Terminal.