Rent is going to fall, economists keep telling us—even Jerome Powell. So why hasn’t it, for over a year?

So why are we still waiting for that to happen?

Over the past year, two-thirds of the core CPI increase has come from shelter, Greg McBride, chief financial analyst at Bankrate, wrote recently. “Shelter remains the largest contributor, responsible for more than half of [December’s] increase in the headline CPI and more than two-thirds of the increase in core CPI over the past year.”

On Wednesday, Federal Reserve chair Jerome Powell, fielding multiple questions about housing inflation prices, essentially agreed with McBride, expressing confidence that they would eventually fall. 

“The question is, when will these lower market rents find their ways into measured rents,” he said. “We think that’s coming, and we know it’s coming, it’s just a question of when and how big it will be.” 

But when you talk to more economic experts about exactly when this will happen, they acknowledge that it just hasn’t happened yet, and it’s yet another surprise in a post-pandemic period that has wrongfooted many of those who project such things for a living.

“We are actually frustrated, we thought that home prices would be a much bigger contributor to the downward movement of inflation,” Eugenio Aleman, chief economist at Raymond James, told Fortune. 

And Capital Economics on Wednesday poured cold water on the idea of future rent drops, citing “reliability issues” with how the government measures rent data. “[W]e remain confident that they will flatline in 2024, rather than fall,” property economist Thomas Ryan said in the note. 

Here’s why so many experts have kept expecting shelter inflation to fall—and the rent you pay in your city or town to approach something that seems reasonable again—and why they just don’t know exactly when that will happen.

The government doesn’t measure housing ‘value’ the way people do

Okay, bear with us here: This is a bit wonky.

The way the government measures rents and home prices in the Consumer Price Index, the main metric for consumer inflation, is complicated, to say the least—and it diverges significantly from how regular people interact with those numbers. And it actually has much more to do with rent prices and for-sale home prices.

Because the CPI aims to illustrate the change in the prices of goods and services over time, the Bureau of Labor statistics tracks the consumption value of a home instead of its outright price. Rather than directly considering the price a home is listed at, the BLS tries to consider what the homeowner would be paying if they were renting, not owning.

The BLS calls this “owner’s equivalent rent,” or the amount of rent that a homeowner would be paid if they were renting out their house, and gathers this data via a questionnaire.

“The owners’ equivalent rent, which is normally what drives the biggest chunk [of shelter costs], is basically a survey. They go to individuals and say, ‘how much would you rent your home for if you had to rent your home,’” says Eugenio Aleman, chief economist at Raymond James. 

“This is a component of the [consumer price] index that is very important, but it’s not highly reflective of what’s actually happening in the market,” he notes. 

Despite how “esoteric” this measurement is, Aleman suggested that the BLS can’t quickly change how it calculates house prices, because doing so would make it impossible to compare home prices over time. “The measures don’t match,” he said.

When it comes to actual rents, meanwhile, the government’s measurement has the effect of spreading out sharp rent increases over 12 months, or more, to reflect for the fact that leases usually get renegotiated only once a year. 

“You have to remember when home prices go up, they don’t impact everyone, they only impact the movers, the buyers. Same things with rent,” Selma Hepp, chief economist at CoreLogic, told Fortune. “It’s not like the cost of eggs, or gas, that you need on a regular basis.”

“You really, on a monthly basis, get only 1/12 of the change that you’re seeing in the market play out in the CPI,” she added.

It’s a sticky situation

Since it takes so long to gather rent information, we have something counterintuitive: Housing inflation rates, the very things designed to communicate how expensive housing is, just don’t always accurately indicate how expensive housing is—or isn’t. When it comes to housing inflation, the experts are now struggling for the CPI data to catch up with the real world.

“The way in which housing or shelter enters the CPI makes it a very sticky measure, and also makes it really hard to know how it might change in the months ahead,” Lisa Sturtevant, chief economist at Bright MLS, told Fortune.  

Private listing services, like Rent.com or Zillow, monitor the listed prices of housing and track their changes in real time. The CPI data, however, has a serious lag—Sturtevant said it can take anywhere from six to 18 months for real-world conditions to be reflected in CPI.

This is to compensate for the irregular way that most people experience rent hikes–fairly high increases, happening rarely. 

“If you are renting an apartment and you go to your landlord who is interested in raising rents, he or she can’t raise your rent because you’re probably on a lease,” she explained. “So it’s going to take another six months when you renew your lease to see those rent increases in the calculation and the data that they collect.”

To make up for this, Bureau of Labor Statistics workers only collect rent data twice a year, with the specific months depending on the particular data sub-sample. This has the effect of smoothing out sharp rent increases—but it means that changes on the ground take a long time to show up in government data.RIght now,

“If I was out there in the rental market right now, I definitely would not use the CPI shelterfilter measure to tell me where rents are headed, because it’s telling me where rents were six months ago,” Sturtevant said.

Location, location, location

The other issue with using CPI measures as an indicator for making sweeping statements about the housing market is that, at its core, housing is a localized issue. 

“CPI, by measure, is intended to give a national picture of what’s going on in the economy,” Sturtevant says. “Housing is local. And so whether the shelter measure is coming up or down, really doesn’t have anything to do with a consumer who may be in Baton Rouge or New York looking to rent a place. Local market conditions are what’s most important.”

Instead, she suggests that consumers focus more on their local markets instead of a generalized, national figure. Basically, she says, ignore what the CPI tells you (in this context).

“As folks are looking to get intelligence on where home prices and rents are going, that I would argue that the shelter component of CPI is perhaps not a very helpful measure to them,” Sturtevant says. “Looking at local trends in rents is really the most valuable thing for a consumer in the market for renting or buying.”