A big difference is what goes into the baskets of goods and services each index tries to measure. The CPI focuses on what consumers pay directly. The PCE is broader. For example, the PCE would factor in the cost of medical care paid by employee-sponsored health plans, Medicare and Medicaid, while the CPI would only take into account out-of-pocket spending by consumers themselves.
The two emphasize different factors, says Ryan Sweet, director of real time economics at Moody’s Analytics. The PCE weights medical care spending while the CPI puts more weight on shelter costs. The PCE measures rural and urban spending, he added, while the CPI just looks at urban. And the PCE includes spending by non-profits not just consumers.
The PCE also factors in shifts in consumer behavior, for example, when people substitute cheaper items for more expensive ones. So, if the price of steak skyrockets and people start buying more chicken and less steak, the PCE’s basket of goods would shift to reflect that, while the CPI’s basket would remain the same as before.
The Bureau of Labor Statistics calculates the CPI using household survey data. The Bureau of Economic Analysis comes up with the PCE using the same data that goes into the quarterly gross domestic product report plus information from a variety of business reports, and it can revise that number just like GDP gets revised.
The Cleveland Fed says CPI has run about a half a percentage point higher than PCE since 2000, though only about 0.3 percentage points higher since 2008.
As for why any of it matters, the Fed has been trying to raise rates from historic lows without derailing an economic expansion. It wants to keep inflation in check without setting the economy back, often a tough line to walk.
“Going forward, data on inflation will be critical for the Fed” as it decides whether and when to raise rates again, Sweet said.
Be the first to comment