People tend to be surprised, and sometimes upset, when they find out that interest expenses aren’t included in the Consumer Price Index. It seems only natural that if people are paying more interest because rates have gone up, it should be reflected in the official measure of inflation.
Matt Stoller, the research director for the American Economic Liberties Project, who has a channel on Substack, claimed that interest was excluded in the 1980s to make inflation appear lower. “The people who originally broke the tools for political purposes aren’t there anymore,” he wrote, and “today’s political class doesn’t even know what they don’t know.”
The government’s explanation for why it doesn’t count rising interest expenses as a form of inflation is subtle and won’t satisfy everybody, but it does make sense when you stop and think about it. So this newsletter will be a partial defense of the government’s way of doing things.
First, though, I want to agree that the pain of paying higher rates, whether for an auto loan or a mortgage or a credit card, is real. In fact, the Federal Reserve counts on that pain. Higher rates help the Fed reduce inflation by forcing consumers to cut back on their spending.
From March 2022, when the Fed began raising rates, through this July, personal interest payments, excluding those on mortgage loans, rose 69 percent, according to Bureau of Economic Analysis data. The average payment on new mortgages rose 19 percent in the 12 months through July, even though house prices rose only about 3 percent, according to Redfin, an online brokerage.
So why not count those payments in the inflation rate? That’s the way it used to be done. Mortgage interest was in the C.P.I. until 1983. But the logic of having it in there was always sketchy, and at a time when mortgage rates got as high as 18 percent, interest costs were having a huge effect on measured inflation.
“A growing number of concerned parties feel that this component is seriously flawed and that changes must be made in order to maintain public confidence in the index,” Robert Gillingham and Walter Lane, who worked for the Bureau of Labor Statistics, wrote in 1982 in the bureau’s Monthly Labor Review. “Unfortunately,” they wrote, “the current treatment of homeownership has no clear conceptual rationale.”
One problem they identified was that the bureau’s measure didn’t distinguish between a house as an investment and a house as something a person “consumes,” or uses for shelter. If you rent and your rent goes up, that’s just plain bad. But if you own and the price of your house goes up, that’s kind of good. Because a house is partly an investment, the interest you pay to buy it can be thought of like the interest you pay on a margin loan to buy shares of stock.
To put it another way, interest is fundamentally different from the goods and services in the Bureau of Labor Statistics’ marketbasket. It isn’t the price of something you consume. It’s the price you pay for shifting the timing of your consumption. Taking out a 30-year mortgage loan allows you to move into a house now. Without it, you would have to save up to pay in cash, which would take years. Later, paying off the loan will force you to curtail future consumption.
To eliminate the confusion between investment and consumption, the Bureau of Labor Statistics decided to regard the price of owner-occupied housing as whatever a person would have to pay to rent that property. That’s a pure measure of consumption, which is what the C.P.I. is supposed to capture. “Interest rates are specifically out of scope” for the C.P.I., Steve Reed, an economist in the C.P.I. branch of the bureau, told me.
Interest rates do affect the C.P.I. indirectly, a point that critics don’t usually recognize. On the demand side, if the cost of owning a home goes up because of rising interest rates, that will push up rents as well, because more people will choose to keep renting. (And rents, again, are the basis for measuring shelter inflation.) On the supply side, owners of apartment buildings will want to charge higher rents because alternative investments, such as Treasury bonds, become more lucrative when interest rates go up.
Of course a house isn’t the only thing people buy that is consumed over time. Autos are the second-biggest example. And in fact the bureau went on to eliminate interest on auto loans from the C.P.I. calculation in 1998.
Credit card interest is excluded from the C.P.I. calculation by a similar logic. The interest is the price of shifting the timing of consumption, not an expenditure for actual consumption, Marshall Reinsdorf, an independent economic consultant who has worked for the Bureau of Labor Statistics and other agencies, told me.
That’s the big picture. But there are some complications, which is why I said my defense of the government’s approach is partial. One complication is whether the C.P.I. is properly regarded as a measure purely of the prices of the goods and services that people buy, or as a measure of the cost of living more broadly. If it’s a pure measure of the cost of goods and services, then interest expense clearly doesn’t belong. If it’s a measure of the cost of living, things are cloudier. Reinsdorf said that he favors keeping the C.P.I. as is, but that interest expense could be taken into account in an alternate version for research purposes.
Reed, the B.L.S. economist, said that interest isn’t in the C.P.I. because the C.P.I. is focused on the price of current consumption, while interest rates reflect the trade-off between current and future consumption. As a practical matter, though, given that so many consumers choose to pay for current consumption in the future via loans or credit cards, when interest rates go up, people who owe money are worse off — their cost of living has risen.
“One can make the argument, yes,” he said. “It’s not a completely ridiculous argument that when interest rates get higher, that affects how well off consumers are. So it’s not a completely cut and dried issue, I agree.”
The Readers Write
Your newsletter on moral luck helped me to better understand what happened to me as a combatant senior leader in the U.S. Army. I served four consecutive combat tours in Iraq and Afghanistan. I operated in a gray moral area exemplified by Clausewitz’s observation that war is the most extreme reality when it comes to actions being decided by being at the right or wrong place at the right or wrong time. It is a significant part of what led to my being diagnosed with severe combat PTSD, of a specific subtype defined as “moral injury.”
Your article on the luck of being good gave me an “ah-ha” experience. And a healing one, too. I have toiled over the years with believing that I was personally responsible for my decisions in war. This is what the Marines and the Army indoctrinated me to believe. Or, was I? Was it just the luck of the draw? Or, did I just say, “screw it,” and put my head down and just keep going. I just couldn’t make sense of it. The whole swirling moral mess just ate me alive.
Your article helped me to make better sense of it, to understand it, to maybe help me to truthfully forgive myself for what I thought were good actions that led to unintended deleterious consequences for others that I could never have predicted. Especially to those who were my subordinates for whom I cared for deeply and who carried out my orders with the best of intentions.
Col. Jeffrey Stolrow, retired
Newport Beach, Calif.
Quote of the Day
“Pet stores do not sell bicycles and bicycle shops do not sell pets, but that fact alone hardly suggests that they have agreed to stay out of one another’s markets.”
— Herbert Hovenkamp, “The Rule of Reason,” Florida Law Review (January 2018)