# Inflation Measures

“Everyone is entitled to his own opinion but not to his own facts.” – Sen. Daniel Patrick Moynihan

September 30, 2018

by Steve Stofka

The above quote has been attributed to the former Senate Majority Leader. People repeat the quote when discussing a contentious subject. We are often convinced that we have the facts when our facts may indeed be arbitrary. Let’s take the case of real or inflation-adjusted income. Has the average real wage declined or risen in the past decades? The calculation depends on which measure of inflation we choose.

There are two measures of inflation, the Consumer Price Index (CPI) and the Personal Consumption Expenditure Price Index (PCE). The CPI relies on surveys of what consumers buy. The PCE is based on surveys of what businesses sell (Note #1). The CPI uses a fixed basket of goods, regardless of changes in the prices of items in a basket. If the weekly basket of goods includes two pounds of ground beef, that two pounds never changes in response to lower prices. It is static. The PCE does adjust for price changes. If the price of a pound of ground beef went down thirty cents, the PCE calculates that a family bought a bit more ground beef and a little bit less chicken, for example. It is a dynamic measure.

People drive fewer miles and buy more fuel-efficient cars as the price of gas increases BUT only after a certain dollar amount. Our purchasing patterns are both static and dynamic. Because we are creatures of habit, our buying patterns are resistant to change. Within a certain price range, we will continue to buy the same items. Outside of that range, we do make changes because we want to optimize our choices.

In the past forty years the CPI has calculated an annual rate of inflation that is over ½% higher than the PCE rate. That small difference compounded over forty years amounts to 23%. That large difference tells two very different stories. Using the CPI, the average worker has lost a few percent in inflation adjusted hourly wages. Using the PCE, on the other hand, the average worker has enjoyed real gains of 20% in the past forty years (Note #2).

Our most volatile disagreements are in areas where facts are difficult to observe. The household survey data that underlies the CPI is unreliable because people living busy lives are not accurate journal keepers of their daily purchases. On the other hand, surveys based on business sales are inaccurate because people stock up on items whose prices decline.

Even when facts are readily verifiable, the interpretation of those facts varies with context. In arriving at our version of the meaning of those facts and their context, we subtract a lot of observable data.  We must filter reality because we cannot manage such a large amount of information. Because we filter our perceptions, eyewitness testimony is unreliable. Although our perceptions are inaccurate, we must act on those perceptions and hope that they are accurate enough. That same reasoning guides economists, politicians, and those in the social and physical sciences. We would all have more constructive discussions if we understood the imperfection of our perceptions.

//////////////

Notes:

1. The Difference between CPI and PCE {Federal Reserve}

2. Using the average hourly wage for production and non-supervisory employees.