On Tuesday, the Bureau of Labor Statistics released inflation data for April, and the headline figure is that consumer prices fell 0.8 percent from a month earlier. Not that they fell at an annualized rate of 0.8 percent, but that they fell 0.8 percent month over month — equivalent to an annualized inflation rate close to minus-10 percent. That was the steepest monthly decline since December 2008.
Still, you should take that figure with a grain of salt. Nearly everything in the world is super-weird and disrupted right now, and the tabulation of inflation statistics is no exception.
Each month, the federal government collects data about prices to see how they have changed. Then the BLS weights that price-change data based on assumptions about how much consumers spend on various kinds of goods and services: If consumers typically devote 14 percent of their spending to food, then the change in food prices makes up 14 percent of the overall inflation statistic. Within the food category, price changes are weighted based on how much consumers typically spend on which kinds of foods, and so forth. The problem there is the word “typically”: Consumer-spending behavior in April was far from typical, and certain things that fell sharply in price — like gasoline, men’s suits, and airfares — also fell sharply in quantity of consumption. As such, the inflation statistic is likely materially lower than it really should be, if you’re trying to measure prices as experienced by consumers.
“The number has to be interpreted thoughtfully and cautiously,” says Stephen Reed, an economist at the Bureau of Labor Statistics. A separate survey — the Consumer Expenditure Survey — is used to develop the weights reflecting how much we spend on what things, and the weights are changed only every two years. Usually that’s fine because the month-to-month changes in the makeup of consumer expenditure aren’t big enough to matter very much (and seasonal adjustments are used to account for changes that occur cyclically every year), but this April has been more different than usual from previous months.
If you look through the CPI release for the price categories that fell sharply, what you’ll find is a list of things you’re probably not doing much of anymore. Retail gasoline prices fell 21 percent last month as people worldwide stopped driving. Various kinds of travel services became a lot cheaper: airfares down 15 percent, vehicle rentals down 17 percent, “other lodging away from home” (read: hotel and motel rooms) down 8 percent. And during a month when Americans were far more likely than usual to sit around their homes in old sweatpants, apparel prices fell 5 percent. (All these price changes are seasonally adjusted.)
Prices have not fallen hard in every industry where sales volume is far down. Dental services did not change sharply in price, because dental offices have not been striving to fill empty dental chairs; more often, they have just closed. But where business must go on, prices have fallen. Airlines have legal mandates to keep operating; oil producers have wells they can’t easily shut off; and hotels (especially at the low end) might as well keep their doors open with skeleton staffing. In all three cases, the industries have used price cuts to move inventory.
While consumers are not enjoying the benefit of falling airfares, they are bearing the brunt of price increases on key components of the CPI. Most notably, prices for food at home rose 2.6 percent in April. Some kinds of foods had particularly large spikes in price: Eggs were up 16 percent, chicken 6 percent, and pork chops 7 percent. But even food categories not affected by coronavirus-related disruptions of meat- and poultry-processing plants tended to show at least modest price increases. Household paper-product prices rose 5 percent.
Food prices, like fuel prices, tend to be sensitive to shocks on the supply side, which is a reason why monetary policy-makers tend to look at a “core inflation” statistic that excludes food- and fuel-price changes when deciding how to set interest rates. But food and fuel prices obviously matter to consumers like any other prices, and those food-price spikes — coming at a time when the supermarket is one of the few places consumers continue to spend heavily — mean the average consumer is unlikely to be experiencing the kind of negative inflation the data appear to show.