This month’s CPI report was firm, although broadly consistent with the signals coming from the distributional measures. Over recent months, we had highlighted the possibility that inflation readings could come in somewhat softer, and this has indeed materialized considering especially April’s reading. That said, in the near-term we now expect readings on average consistent with the 2% target.
More broadly, we continue to believe that the Japanese inflation process remains fundamentally different from the pre-Covid period. In particular, the signals from our models suggest that inflation is still running around target—perhaps slightly below it, but nevertheless relatively close to the target level. Looking ahead, inflation is likely to remain broadly around target. That said, a modest pass-through from higher energy prices to core inflation cannot be ruled out. We estimate this effect to be limited, on the order of one to two tenths of a percentage point with some upside risks.
As for the Bank of Japan, our impression is that the BOJ has finally moved interest rates into more reasonable territory. Current policy rates are more consistent with the prevailing level of inflation than they were in the past. Further tightening remains possible, although we would currently place the upper bound at roughly 50 basis points of cumulative additional rate increases. That said, we would not be surprised if the BOJ decided to take more time before proceeding further. The reason is straightforward. Our models indicate that, in 2026, the BOJ staff may be forced to revise down their inflation projections for that year, in line with incoming data. By contrast, our model forecasts for 2027 and 2028 are now broadly in line with, or slightly above, the BOJ staff projections.
A PDF containing all relevant CPI charts can be downloaded here.
A PDF containing all relevant labor market charts can be downloaded here.
EVIDENCE FROM THE MODELS
DISTRIBUTIONS
The overall shape of the distribution continues to point to underlying inflationary pressures that remain close to target. MoM readings are likely to stay in line with recent trends in the near-term.
Figure 1a. Cumulative density function (and table stats) of core-core CPI cross-section price changes.


Figure 1b. Kernel of core-core CPI cross-section price changes.

Figure 1c. Median of core-core CPI cross-section price changes.

Proxies of BoJ’s underlying inflation
Figure 2 illustrates the three “underlying inflation” measures published by the BoJ (blue lines). For each, we’ve calculated a proxy (yellow lines) based on the distribution of price changes.
The key takeaway: The proxies are going sideways. In any case, they all remain well above the pre-Covid levels.
Figure 2. BoJ measures of underlying inflation and our proxies (%).



MEDIUM-TERM FORECAST
Today’s data had little implications for the model-based forecasts, as our Q2 nowcast is basically unchanged. For core-core, the forecast remains above target at the end of the forecast horizon.
Figure 3a. “Main” Phillips curve model forecast, Index ex fresh food (BoJ) (YoY, %).


Figure 3b. “Main” Phillips curve model forecast, Index ex fresh food and energy (core-core) (YoY, %).


TREND INFLATION
Figure 4 presents our trend inflation models for Japan, based on Rudd (2020). The sample includes now Q2. These models suggest that trend inflation is at 1.6-1.7%.
Taking into account inflation expectations and general equilibrium dynamics (particularly wages), we estimate pi* for Japan to be around the BoJ’s 2% target*.
Figure 4. Trend inflation.

COMPARISON WITH BOJ FORECAST
Figure 5 compares our model-based forecast with the BoJ’s latest projections. The table also includes estimates for “acquired inflation” and “carry-over” effects. For FY2026 we estimate that acquired inflation is 0.4% (NSA) for BoJ core and 2.8% (NSA) for core-core. The models now point to some downside risks for FY2026, as the data have been -on net- on the soft side recently. Having said that, the models remain above target going forward.
Figure 5. Underlying Inflation forecast vs BoJ forecast vs “acquired inflation”
