Japan core CPI (excl. fresh food) year-over-year inflation, May 2026

Direct realized signal is limited (Mar 1.8 -> Apr 1.4, a -0.4pp move driven by the subsidy step). Japan core is normally smooth (typical monthly YoY moves +/-0.1-0.2pp); the -0.4 was a one-off policy/base shift. Absent another subsidy step, a flat-to-slightly-lower May near 1.3-1.5 is most likely. I set the 80% half-width at ~0.4pp to reflect policy-driven jump risk rather than the tiny ordinary-noise stdev.

Reference class: Japan core CPI monthly YoY changes are among the lowest-variance in the G7 in calm periods (mostly within +/-0.2pp), but subsidy on/off events produce occasional 0.3-0.5pp jumps. Conditioning on an active-subsidy regime, the modal move is small; the tails are driven by subsidy policy changes.

With subsidies still in place and core-core softening, core most likely holds near the April 1.4% level or drifts marginally lower. Point estimate 1.4%.

Lands outside [1.0, 1.8] if: subsidies are scaled back or expire and oil passes through, lifting core toward ~1.9-2.0 (upside tail), or an expanded subsidy / sharp energy drop pulls it below ~1.0 (downside tail). Food (rice) and services pass-through could also surprise to the upside. Because the dominant variable is discretionary subsidy policy, the distribution is bimodal-leaning rather than smooth; the interval captures the central mass but a policy shift could exit it.

The resolver is the Statistics Bureau of Japan/e-Stat first print for the national CPI subgroup 'all items less fresh food' for May 2026, reported as the year-over-year percent change rounded to one decimal.

Base-rate/reference-class: monthly Japan core CPI excluding fresh food usually moves gradually at one-decimal precision; a one-month change of roughly 0.0 to 0.2 percentage points from the latest national print is more likely than a large break without a tax, subsidy, or energy shock.

Anchor at the latest national readings: average of 2026-02 to 2026-04 is (2.9 + 3.0 + 3.1) / 3 = 3.0. The Tokyo May preliminary at 2.8 suggests slight downside versus April, but national core tends to be less volatile, so I keep the point at 3.0 and set an 80% interval of 2.5 to 3.5.

Counter-consideration: imported energy, yen depreciation, administered-price changes, or food-related pass-through outside fresh food could keep May national core closer to or above April's 3.1 rather than easing with Tokyo.