The outlook for inflation in the next couple of years may depend on when America's renters choose to move. A new report points to rents remaining a stubborn obstacle that prevents overall inflation from returning durably to 2%.

Why it matters: Rent is the key reason inflation remains elevated in key government data, even as private sector sources have pointed to a stabilization in rents.

• A Cleveland Fed paper provides some potential answers for what's going on.

The intrigue: There is an important divide in renters — those who have moved in recent years, signing a new lease with higher rents, and others who have stayed in place with ongoing, existing leases.

• The "rent gap" between these two groups — that is, the difference between current rent and what might be paid by a new tenant — remains unusually large, according to the Cleveland Fed's model.

What they're saying: "When new-tenant rent inflation has been high, continuing-tenant rents have risen more slowly, in effect providing the latter group a discount to remain in their units," Cleveland Fed economists Lara Loewenstein, Jason Meyer and Randal Verbrugge write.

By the numbers: Before the pandemic, the researchers estimate the rent gap was about 1% — meaning that, on average, existing tenants paid about 1% less in rent than what a landlord might command for a new tenant.

• As of last month, this gap is hovering around 5.5% — well below the 11% peak but notably above pre-pandemic norms.
• That elevated gap suggests that when renters move, higher rent likely awaits them, according to the researchers.

The big picture: The Cleveland Fed model says this lagged passthrough means the Consumer Price Index's shelter category — the largest component of the inflation gauge — will remain above pre-pandemic levels until mid-2026.

• If the researchers are right, that could mean an extended period of upward pressure on overall inflation from shelter alone (and that doesn't take account of any other concurring shocks that might push up other categories).

What to watch: That forecast is highly uncertain. The researchers assume that trends around mobility and the extent to which landlords hike rent on lease renewers and new renters.

• In the early 2000s, roughly 31% of renters moved in the prior year. That has dropped to 22% today, according to Census data. The Cleveland Fed model assumes that continues.
• If renters stay in their units longer or landlords don't raise rents like researchers assume, that could bring the CPI shelter index "below its pre-pandemic average much more rapidly."

For much of the post-pandemic expansion, construction of multifamily housing — apartments, condos and such — has been gangbusters while single-family home construction was more limited. Now the trends are reversing.

Driving the news: New Census data out this morning showed that fewer housing units were permitted and started in September than August. But it is the longer-term numbers that are more interesting.

• Compared with a year ago, the number of single-family homes that started construction is up 5.5%, according to this morning's data. That bodes well for supply in a slice of the housing market that overwhelmingly serves home buyers, rather than renters.
• Meanwhile, the number of homes started in projects containing five or more units, which disproportionately serves the rental market, is down 15.7%.

Between the lines: Builders are grappling with a higher cost of borrowing and the impact of new competition due to all the rental projects started from 2021 to 2023 delivering. That is making it harder for new multifamily projects to "pencil out," or make financial sense.

• It is one more example of a bullwhip effect, in which the seismic changes in supply and demand for all sorts of goods unleashed by the pandemic are still rippling through the economy.
• Separate from the rental dynamics identified by the Cleveland Fed in the piece discussed above, falling multifamily construction could be a source of upward pressure on market rents in the next couple of years.