Source: Federal Reserve Bank of Cleveland via FRED
Data through: September 2026 · updated Oct 15 · Updates: Monthly · Next: Nov 10
MThe three bars show the last 3 months for this indicator, oldest left to newest right. Bar height reflects each reading relative to the other two — the tallest is the highest of the three, the shortest the lowest.
Historical context: Currently in the 34th percentile of all historical readings.
What this means right now: Median CPI is near the Fed's 2% target — between 1.5% and 3.5%, as in about 7 months in 10 since 1983 (355 of 514). Most of the CPI basket is seeing steady price growth.
Median CPI · Monthly · 1982–2026
NBER recessions
Scroll to read · zoom and pan in Expand
Monthly values
Median CPI — the last 12 months
Every month's reading, with the band the heatmap gives it. The chart above shows the same series in full.
Month
Value
Band
Sep 2026
+2.6% YoY
Within normal range
Aug 2026
+2.6% YoY
Within normal range
Jul 2026
+2.7% YoY
Within normal range
Jun 2026
+2.7% YoY
Within normal range
May 2026
+2.9% YoY
Within normal range
Apr 2026
+2.8% YoY
Within normal range
Month-end readings of the same series the chart shows, from median_cpi_yoy_12m. Values are the current record, not point-in-time: a month that has since been revised shows its revised value.
What is the Median CPI?
The Cleveland Fed Median CPI is a measure of core consumer inflation that takes a different approach from CPI or Core PCE. Each month it finds the median price change across all CPI components — the item sitting exactly in the middle of the distribution. This automatically strips out whatever the extreme movers happen to be that month, whether a supply-chain shock, a volatile energy category, or a one-time base effect.
Where Core PCE excludes food and energy by category, Median CPI excludes whatever the outliers happen to be — a different and complementary noise filter. On the site's own data since 1984 it has been a better guide to inflation a year or more ahead than headline CPI at every horizon, and about level with core CPI. As a guide to core PCE a year or more ahead, core PCE itself has done better. That is the Cleveland Fed's case for it: it keeps the persistent part of inflation and drops the noisy tails.
The result is a clean read on sticky, broad-based inflation — the underlying trend behind the headline. A Median CPI running above 3.5% means the majority of the CPI basket is seeing elevated price growth, not just a few hot categories. A reading near 2% indicates broad-based price stability.
We display the 12-month change. The Cleveland Fed also publishes a monthly annualized version, but it is much noisier; the 12-month rate is the cleaner read on persistence. The middle reading of this series since 1983 is about 2.85% — a little above the Fed's 2% PCE target, because CPI runs higher than PCE.
Around 2% is the healthy zone, the CPI equivalent of the Fed's 2% PCE target. Both persistently high and persistently low inflation are treated as problems: the colour bands turn red at both ends.
How We Color-Code the Median CPI
Our heatmap colors this indicator against historically significant thresholds:
Below 1.0%
Deflation risk — below 1%, seen in only 12 months since 1983, February 2010 to January 2011
1.0% – 1.5%
Below target — 1% to 1.5%; inflation drifting low, as in only 8 months since 1983 (2009–11)
1.5% – 3.5%
Healthy — 1.5% to 3.5%, near the Fed's 2% target; the series' middle reading is about 2.85%, and about 7 months in 10 since 1983 sit here (355 of 514)
3.5% – 4.5%
Warm — 3.5% to 4.5%, above the comfort zone; about 1 month in 5 since 1983 (94 of 514), in 1983–91, 2021–22 and 2024–25
Above 4.5%
Hot — above 4.5%, with most of the CPI basket seeing elevated price rises; 45 months since 1983, in 1989–91 and 2022–24
Historical Extremes — What Happened Next?
When this indicator reaches extreme levels, history shows consistent patterns:
Feb 2023Post-COVID inflation peak (sticky/late)
+6.9% YoY
Median CPI peaked 8 months after headline CPI — its hallmark, since broad-based inflation is slow to recede. By February 2023 headline CPI had fallen to 6.0% from its 9.0% peak, but the median showed price rises still spread across most of the basket.
Dec 2022Height of the 2021-23 inflation surge
+6.8% YoY
Broad price pressure across most of the CPI basket — not just a few energy or shelter categories. The median showed the inflation was widespread, not a quirk of a few outlier components.
Aug 2010Post-GFC disinflation trough
+0.5% YoY
The series' low — the post-crisis deflation scare, in the months before the Fed launched QE2. Broad-based price growth had all but stalled across the consumer basket.
Aug 1990Pre-recession inflation peak (Gulf War era)
+4.8% YoY
The high of the early-1990s inflation episode, as oil prices surged at the start of the Gulf War. The 1990–91 recession followed, and Median CPI fell through the 1990s, from 4.6% on average in 1990 to 2.3% in 1999.
Frequently Asked Questions
How is Median CPI different from CPI or Core PCE?
Headline CPI averages all price changes weighted by spending. Core PCE excludes food and energy by category to focus on underlying inflation. Median CPI takes a different approach: each month it finds the price change sitting in the exact middle of the distribution of all CPI components. That automatically excludes whatever the outliers happen to be that month — energy, rents, or an unusual base effect in any category. This makes it less dependent on which categories you choose to exclude and more responsive to breadth of price changes. If 51% of the basket is seeing price gains above 3%, the median will show above 3%, regardless of which specific categories are hot.
Why use a median instead of an average or a category exclusion?
A simple average of price changes can be moved by a single extreme category — a 40% jump in used-car prices lifts the average even if everything else is stable. The median is not: one component going to an extreme does not move the middle of the distribution. As a guide to inflation a year or more ahead, the median CPI has beaten headline CPI at every horizon on the site's data since 1984. Two years out it missed headline CPI by 1.3 points on average, against headline's own 1.6, and it edged core CPI at two and three years. It has not beaten core PCE as a guide to core PCE. Bryan and Cecchetti's 1992 Cleveland Fed paper made the case for it.
Why are both high AND low readings concerning?
The Fed's mandate is price stability — specifically a 2% PCE inflation target (roughly equivalent to about 2.3–2.5% CPI). Inflation well above target (Median CPI above 4.5%) means broad erosion of purchasing power. But inflation persistently below 1% carries deflation risk. When prices fall broadly, consumers delay purchases (expecting lower prices tomorrow), businesses cut investment, debt becomes more expensive in real terms, and the Fed has limited room to stimulate. Japan's "Lost Decades" are the canonical example of deflation's self-reinforcing damage. The Median CPI's U-shaped color scheme reflects that both extremes are harmful.
Why display the 12-month YoY and not the monthly number?
The Cleveland Fed also publishes MEDCPIM158SFRBCLE, the monthly-annualized version of Median CPI (what the 12-month change would be if that month's change were repeated for a year). That series is significantly noisier — a single month of unusual seasonal patterns or data revisions can swing it dramatically. The 12-month YoY smooths over those fluctuations and shows the underlying trend in persistent inflation. For reading the broad inflation backdrop, the 12-month rate is the steadier guide; the monthly annualized figure is the noise.
Why is the Median CPI historical median around 2.85% if the Fed targets 2%?
The Fed's 2% target is specified in terms of the PCE deflator, which systematically runs lower than CPI by roughly 0.3–0.6 percentage points. This gap exists because PCE uses different weighting (chain-weighted, adjusting for consumer substitution) and covers a slightly different basket (including healthcare paid by employers and government). Median CPI is a CPI-based measure, so its "neutral" level is inherently slightly above 2% when translated to PCE terms. Additionally, Median CPI's long-run historical median of ~2.85% reflects the full 1982–present sample, which includes the elevated inflation of the 1980s-early 1990s. Since the mid-1990s, the typical range has been closer to 2–3%.
How are this page's colour levels set?
Four levels set the colours. 1.5% and 3.5% bound the healthy range, around the Fed's 2% target; a CPI-based gauge sits a little above the PCE measure the Fed targets. Above 3.5% inflation is warm, above the comfort zone; above 4.5% it is hot, with most of the CPI basket seeing elevated price rises. On the low side, below 1.5% inflation has drifted below target, and below 1% deflation is a risk. Both ends are coloured as problems, because persistently high and persistently low inflation both do damage. They are fixed levels, set in June 2026 with the series' whole history in view, and they do not move as new data arrives. A reading above 4.5% gets the same colour in 1990 as today, so a change of colour is a change in median inflation, never in the yardstick.