The Consumer Price Index, or CPI, tracks how the cost of daily life changes over time. Shelter is one of its biggest pieces. Haver Analytics puts its relative importance at 35% of the index. That is an analyst figure, not a number the BLS pages supplied here confirm. Even so, a category that large moves the headline inflation rate when it moves.
What shelter covers, and what we could not confirm
The BLS groups shelter into rent of primary residence, owners' equivalent rent (OER), and lodging away from home. The source pages supplied for this piece did not include the BLS text on those components or their item weights. Treat that list as a general description and check the BLS pages for the exact weights.
The idea behind OER is simple. Homeowners do not pay rent to themselves, so the index estimates what a comparable home would rent for. The BLS factsheet on rent and rental equivalence is the place to read the official method. Its body text was not available to us, so we do not describe its details here.
One consequence matters. The shelter index tracks the cost of housing services. It does not track house prices or mortgage interest. A Brookings paper on housing policy and inflation argues that shelter inflation may overstate the importance of rents relative to house prices and mortgage costs, which shape the cost of owning a home but do not directly enter the measure.
How the six-panel sample creates timing effects
The index does not survey every home every month. Per Haver's description of the BLS method, housing units are split into six rotating panels. The monthly percent change in shelter is the sixth root of the percent change over the previous six months for one of those panels.
That design smooths the series. It also means each reading leans on whichever panel came up that month. A panel is revisited only every six months, so a missed collection can leave a mark that lasts until the panel is surveyed again. The index is a set of rolling samples. It is not a live picture of the housing market.
The October 2025 gap and the expected correction
The partial government shutdown of October 2025 showed how this works. Haver says the BLS did not collect the monthly survey, assumed October shelter prices equaled April's, and calculated a zero monthly change. The understatement would persist until that panel was surveyed again in April.
Haver's illustration uses numbers. The six-month shelter change was 1.65%, not annualized, in September 2025. With October treated as unchanged, Haver says the shelter index was low by about 0.3% and the overall CPI by about 0.1%. Any change spanning October, including annualized ones, missed a month of shelter inflation.
Haver expected the correction to add roughly 0.1% to both the monthly and 12-month CPI change when the April figures were released on May 12. The sources here do not confirm whether that happened as expected, and they give no official revision history beyond Haver's account. Check the BLS release for the actual result.
What the recent index values show
The FRED series for shelter (CUSR0000SAH1) is a seasonally adjusted index with 1982-1984 equal to 100. Its recent monthly values in 2026 are listed below.
- May: 427.998
- June: 428.501
- July: 429.095
- August: 430.227
- September: 431.908
That is a rise of 3.910 points, or about 0.9%, over four months. FRED lists the last update as October 14, 2026 and the next release as November 10, 2026. The BLS also publishes non-seasonally adjusted shelter data. The two series will not match month to month.
How to compare your own rent to the index
The sources supplied did not offer a consumer method, so this is a simple approach of our own, not BLS guidance. Take your rent today and your rent from a year ago. Divide the difference by the older figure. That is your percentage change.
Then compute the same for the index. Divide a recent value by the value twelve months earlier, minus one. Use the same type of series each time, seasonally adjusted or not. If your change is far above or below the index, that says little by itself. The index is a national average built from rolling samples and estimated owner rents.
Brookings also describes housing as a key part of post-COVID inflation, which stayed above the Fed's 2% target, in its piece on quantitative easing and housing inflation. That is context for why the number draws attention. It is not a forecast of your next lease.
