The Consumer Price Index reported 3.2 percent annual inflation in its most recent release. The consumer standing in the grocery checkout line, paying 30 percent more for eggs than two years ago and $200 more per month in rent, experiences a disconnect so severe that it has eroded public trust in the statistical apparatus of the federal government. The disconnect is not evidence of manipulation. It is a consequence of methodology — specifically, the way the Bureau of Labor Statistics measures shelter costs, which comprise more than one-third of the CPI basket and lag real-time market rents by twelve to eighteen months. Understanding that lag, and understanding the alternative inflation measure that Jerome Powell actually watches, is the difference between reading the headline number and understanding what it means.
How the CPI Is Constructed
The Consumer Price Index measures the average change in prices paid by urban consumers for a fixed basket of approximately 80,000 goods and services, weighted by the proportion of household expenditure each category represents. The basket is divided into eight major groups: food and beverages (14.3 percent), housing (34.9 percent), apparel (2.6 percent), transportation (16.7 percent), medical care (8.4 percent), recreation (5.4 percent), education and communication (6.3 percent), and other goods and services (3.2 percent). The weights are derived from the Consumer Expenditure Survey, which the BLS updates every two years.
The BLS collects approximately 94,000 price quotes per month from 23,000 retail establishments and 50,000 housing units. Price collectors visit stores, record prices, and transmit the data to Washington. The data is seasonally adjusted, aggregated, and published on a monthly schedule — typically on the second or third Tuesday of each month, covering the prior month's prices.
The headline CPI — "all items" — includes food and energy. Core CPI excludes food and energy because their prices are volatile (driven by weather, geopolitics, and OPEC decisions) and can obscure the underlying inflation trend. The Federal Reserve's preferred inflation measure is not the CPI at all — it is the Personal Consumption Expenditures price index, published by the Bureau of Economic Analysis, which uses a different weighting methodology that accounts for consumer substitution between goods as prices change.
The Shelter Lag: The Biggest Distortion in American Inflation Data
The single largest source of the gap between reported inflation and experienced inflation is the shelter component — specifically, the "owners' equivalent rent" measure that the BLS uses to capture housing costs for homeowners. OER does not measure what homeowners actually pay (mortgage payments, property taxes, insurance). It measures what homeowners believe their home would rent for in the current market — a hypothetical rental value estimated through a survey of homeowners.
The OER methodology produces a shelter inflation figure that lags real-time market rents by twelve to eighteen months. When market rents surged 15 percent in 2021 and 2022, OER captured the increase gradually through 2023 and into 2024. When market rents decelerated to 3 percent annual growth in late 2023, OER continued reporting elevated shelter inflation — 5 to 6 percent — because it was still absorbing the earlier surge. The lag works in both directions: OER understates shelter inflation during rental booms and overstates it during rental slowdowns.
The practical consequence is that the CPI overstated inflation throughout 2024 and into 2025, because the shelter component was reflecting rental market conditions from twelve to eighteen months prior rather than current conditions. Real-time rent indices — Zillow's Observed Rent Index, Apartment List's national rent report — showed rental inflation at or below 3 percent while the CPI's shelter component reported 5 percent. The 2-percentage-point gap between real-time rents and CPI shelter inflation accounted for the majority of the difference between the headline CPI and the inflation rate that consumers with current leases actually experienced.
What Supercore Is and Why Powell Watches It
Supercore inflation is core services inflation excluding housing — the price change in services such as healthcare, transportation, financial services, education, and personal care, stripped of the shelter component that the lag distorts and the goods and energy components that are volatile. The Federal Reserve does not publish supercore as an official series, but Powell and Fed governors reference it explicitly in testimony and press conferences as the measure most indicative of domestic, demand-driven price pressure.
The logic is straightforward. Goods prices are set in global markets and driven by supply chain conditions that domestic monetary policy cannot directly influence. Energy prices are set by OPEC production decisions and geopolitical disruptions. Shelter prices are measured with a methodology that lags reality by a year or more. What remains — core services ex-housing — reflects the pricing power of domestic service providers, which is driven primarily by wage growth, which is the variable most directly within the Fed's influence through its effect on labor market conditions.
When supercore inflation runs above 4 percent, the Fed interprets it as evidence that the labor market is tight enough to generate wage-price spiral risk — the condition in which workers demand higher wages to keep pace with rising prices, businesses raise prices to cover higher wages, and the cycle feeds on itself. When supercore decelerates toward 2.5 to 3.0 percent, the Fed gains confidence that the labor market is cooling sufficiently to bring overall inflation back to target without requiring further rate increases.
How to Read the Inflation Data
The informed reader of inflation data performs three operations that the financial media typically does not. First, separate the headline CPI into its components and examine the shelter contribution. If shelter is contributing 1.5 to 2.0 percentage points to a 3.2 percent headline reading, and real-time rent indices show rental inflation at 3 percent, the underlying inflation rate is closer to 2.5 percent than the headline suggests.
Second, examine the PCE price index alongside the CPI. Because the PCE uses a different weighting methodology — one that reflects actual consumer spending patterns and adjusts for substitution — it typically runs 0.3 to 0.5 percentage points below the CPI. A CPI reading of 3.2 percent frequently corresponds to a PCE reading of 2.7 to 2.9 percent, which is significantly closer to the Fed's 2 percent target.
Third, track supercore. If supercore is decelerating even while headline CPI remains elevated due to shelter lag, the Fed is likely closer to cutting rates than the headline number implies. If supercore is accelerating while headline CPI decelerates due to falling energy prices, the Fed is likely further from easing than the headline suggests.
The gap between experienced inflation and reported inflation is not a conspiracy. It is a methodological artifact — a consequence of measuring a dynamic economy with statistical tools that were designed for stability. The consumer who understands the tools reads the data differently. The investor who understands the tools reads the Fed differently. And the citizen who understands the tools evaluates the government's economic stewardship on evidence rather than feeling.