Inflation Is Lower Than Most People Think

(Bloomberg Businessweek) -- I got some negative feedback on my Bloomberg Businessweek cover story, “Did Capitalism Kill Inflation?” Lots of readers found it ridiculous that central bankers think inflation is too low. Here’s an excerpt from one email I received from a reader in southwestern Missouri:

I help people with budgets and if you told THEM inflation is dead they would laugh at you and I bet if we looked at YOUR budget we would find plenty of inflation. …
Inflation is the biggest costs consumers face – Rent/Housing, Health Care, Education….
I read your stuff and just laughed, as anyone in flyover country would.
what a bad joke.

This is a widely shared view, not just in what the reader calls flyover country. It is also … wrong. The official measure of inflation, while imperfect, is far more likely to be accurate than the subjective impressions of ordinary consumers. The Bureau of Labor Statistics puts enormous effort into gathering price data for its monthly Consumer Price Index. If you doubt that, peruse its painstakingly detailed, 107-page Handbook of Methods (PDF) or the more reader-friendly FAQ.  

Why, then, is it so common to think that official inflation is understated? Here are a few reasons.

Slow wage growth: Sometimes, when people say inflation is high, what they really mean is that their spending power is diminishing. That’s what you get if inflation is low, but your wage growth is even lower. For most Americans, wage growth has been outpacing inflation—but not by very much.

Loss aversion: People feel losses more acutely than they enjoy gains of the same size. So if you spend more on breakfast cereal but equally less on milk, you may still feel as if you’re coming out behind.

Housing inflation: People often point to the rising prices of houses as evidence of inflation, but that’s not counted in the CPI. This is because the BLS considers a house an asset, like a share of stock, rather than something you consume. Shelter is the biggest component in the CPI, but it’s measured by rents. If you own a house, the “rent” you pay, in the BLS method, is what you could get if you moved out and rented the house to someone else. This has to be estimated. Theoretically, rents should move up and down with house prices, but official inflation won’t reflect a housing price boom if rents are rising more slowly.

Salience and frequency: The price of gasoline is probably the best-known price in America. It’s posted for everyone to see, and we buy gas often. So when gas goes up, it feels as if inflation is high. (The reverse should be equally true, but see above: Loss Aversion.) In 2015, the private American Institute for Economic Research published an Everyday Price Index, which was weighted toward items that people buy frequently, including food. From 2001 through 2014, it said, the EPI grew at an average annual rate of 3.1 percent, vs. 2.3 percent for the CPI. Interesting. But the CPI is still a superior measure of everything people buy, not just high-frequency items.

Quality improvements: For some categories of items, such as computers, the BLS adjusts prices for quality improvements. If a new computer is faster and has more memory than the one it replaces in the consumer market basket, the BLS will count the improvement as a price decrease, even if the price paid remains the same. Consumers don’t necessarily see things that way.

Headline vs. core: Consumers often get mad when they hear Wall Street economists and Federal Reserve officials talking about core inflation, excluding food and energy, rather than the headline measure, which includes everything in the consumer market basket. They think it means that the pros don’t think food and energy prices matter. That’s not the point. The point is that to get a handle on underlying inflation trends, you want to ignore temporary ups and downs resulting from idiosyncratic factors such as trouble in the Libyan oilfields that temporarily raises the price of gas, or a bumper citrus crop in Brazil that temporarily makes oranges cheaper in the supermarket. In the long run, of course, you do care about food and energy price trends. That’s why the Fed’s inflation target of 2 percent is for the headline number, including food and energy.

The Federal Reserve found in a 2016 study that women have higher inflation perceptions than men, on average. It also found that people in the bottom quarter of incomes have higher inflation perceptions than people in the top quarter.

Given all these factors, it’s surprising how close perceived inflation is to the official numbers. In the University of Michigan Surveys of Consumers for April, the median (i.e., midpoint) forecast for inflation in one year was 2.4 percent, while the average (pulled upward by some high guesses) was 3.1 percent. In contrast, the median March forecast by members of the Federal Reserve’s Federal Open Market Committee for the Fed’s preferred measure of inflation in 2020 was 2.0 percent.

Other surveys show a much bigger gap. A Zogby poll in June 2016 found a median inflation perception that year of 10 percent, vs. actual inflation of about 2 percent. I have no idea why Michigan and Zogby would be so far apart. But going with the Michigan number, one possible reason the official/perceived gap is small is that consumers are quite aware of what the government and the news media are telling them about official inflation. So the estimate they give to pollsters is some kind of average of what they hear from the BLS and what they see—or think they see—with their own eyes.

To contact the editor responsible for this story: Eric Gelman at

©2019 Bloomberg L.P.