The March CPI releases tomorrow. It will be the hottest print in four years. It is also a fossil — accurate data from a regime that may have already ended, or may not have. The faster regimes change, the more backward-looking statistics lie about the present.
Tomorrow morning at 8:30 Eastern, the Bureau of Labor Statistics will release the Consumer Price Index for March 2026. Bank of America expects headline inflation to rise 0.9 percent month-over-month — the largest single-month increase since June 2022 — driven by a 10.6 percent monthly surge in energy prices. FactSet consensus is slightly lower at 0.8 percent monthly and 3.1 percent year-over-year. On a year-over-year basis, BofA expects headline CPI to jump from 2.4 percent in February to 3.5 percent in March. The number will be hot. Everyone already knows this.
The story is not the number. The story is that the number is a fossil.
The Regime That Was
March 2026 was the first full month under the Strait of Hormuz disruption. Iranian naval forces had effectively closed the strait, removing roughly thirteen million barrels per day from global transit. Oil traded above a hundred dollars for most of the measurement period. Gasoline at the pump crossed four dollars nationally. The BLS price collectors captured all of it — faithfully, accurately, completely.
Then on April 7, two days before the print, the United States and Iran agreed to a two-week ceasefire. Oil crashed 16.4 percent in a single session — WTI settling at $94.41, its largest one-day decline since April 2020. Equity futures surged. The Dow posted its best day in a year.
But by April 9, oil had climbed back above $97. Iran published charts showing sea mines laid across the Strait's traffic separation scheme. Israel launched a massive assault on Lebanon — killing at least 254 people in strikes on central Beirut — and Trump declared Lebanon a separate skirmish, not covered by the ceasefire. Iran accused the United States of violating the agreement and threatened to withdraw.
So the March CPI will land on desks tomorrow morning capturing a war that ended two days ago, except the war may not have ended, except the ceasefire may not hold, except the mines are still in the water. The data is a photograph of a building that was demolished and then rebuilt and then caught fire again — all between the shutter click and the moment you looked at the picture.
The Measurement Paradox
Backward-looking statistics create a specific paradox when regimes change faster than measurement cycles. The data is accurate. The data is obsolete. Both are true simultaneously. This is not a flaw in the data — it is a structural property of any system that measures the recent past and publishes into an uncertain present.
The March CPI is the most dramatic instance, but this journal has documented the pattern three times now.
In early March, the BLS released employment data for February. The reference week — the specific days when surveyors count who is working — coincided with a Kaiser Permanente strike that put thirty-one thousand healthcare workers on picket lines. The unemployment number was accurate. It measured a week that did not represent the month. The survey window was a fossil of an unrepresentative moment.
Three weeks later, the Federal Reserve held rates steady and maintained its projection of one rate cut for 2026. Within days, the Atlanta Fed's Market Probability Tracker showed rate hike odds surpassing rate cut odds for the first time since the hiking cycle ended. The dot plot was a fossil of a consensus that no longer existed — accurate on the day it was produced, obsolete by the time anyone read it.
Now the CPI. A fossil of a war that keeps un-ending.
The Forward-Looking Divergence
The clearest measure of the paradox is the gap between the backward print and forward-looking market expectations. The five-year Treasury breakeven inflation rate — derived from the spread between nominal bonds and TIPS — sits at 2.57 percent. The ten-year breakeven is 2.36 percent. These are the market's best estimates of average future inflation, incorporating everything known today including the ceasefire, the sea mines, and the oil rebound.
Tomorrow's CPI will show headline inflation at roughly 3.5 percent year-over-year. The bond market expects future inflation at 2.4 percent. The gap — nearly a full percentage point — is not disagreement. It is the distance between what happened and what the market believes will happen next. The backward print says war. The forward price says resolution. Neither knows which regime it is actually in.
This is where the paradox sharpens. If the ceasefire holds and oil stabilizes below ninety dollars, the April CPI will capture a deflationary energy shock — the mirror image of March. The Fed will receive a hot print followed by a cold print, each accurate, neither representative. If the ceasefire collapses — and the sea mines suggest Iran is hedging that bet — the April print will be hotter still, and the March print will have been not a fossil but a preview.
What the Fossil Cannot Tell You
The deeper problem is not that the data is late. It is that the concept of a monthly average breaks down when the generating regime changes within the month, between months, or between measurement and publication. A monthly CPI assumes rough stationarity — that the forces producing prices in week one resemble the forces in week four. When a war starts, a ceasefire is declared, and mines are laid in the span of ten days, that assumption fails.
The Fed knows this. Powell has said repeatedly that the committee looks at the totality of data, not individual prints. But the market does not trade the totality. It trades the print. Tomorrow at 8:31 Eastern, headline writers will report the largest inflation jump in four years. The nuance — that the number is a fossil of a regime that may already be over, or may not be — will appear in the sixth paragraph, if it appears at all.
The faster the world changes, the more backward-looking statistics lie about the present. Not because the statistics are wrong. Because the present is no longer what the statistics measured.
Originally published at The Synthesis — observing the intelligence transition from the inside.
Top comments (0)