No Land For Frankfurt Doves
The March HICP report suggests that core inflationary pressures remain very strong. Our models suggest that the distribution of price changes continues to shift higher, that the median remains elevated, and that the “common component” across items remains solid. Put it differently, we see virtually no good news in today’s report.
“Acquired” inflation in 2023
Mechanically, today’s report implies that “acquired” inflation in 2023 is 4.0% for core HICP and 4.7% for headline HICP. (reminder: this means that if the MoM (sa) will be 0% in each of the remaining months of 2023, at the end of the year the YoY of core and headline inflation will average 4.0% and 4.7%, respectively). The reader can understand that the ECB staff forecast is too low: we still have 9 reports in 2023, core prices are expanding at about 6% (ar) and acquired inflation is already 4.0% for core and 4.7% for headline against the ECB staff forecasts of 4.6% and 5.3%, respectively.
Evidence from the distributions
The distribution: higher it goes (again!). This month, we got mixed signals from the percentiles (not shown) as the majority moved up, but some down. Nevertheless, a strong signal comes from the Kernel of price changes. Figure 1 shows that the distribution of price changes in the last few months continues to travel to the right (higher) and price dispersion continues to be much higher than pre-Covid. The median of the distribution (Figure 2) remains very elevated and in the range of the last 16 months or so.
Let us be clear. Forget all comments about exclusion indexes. Forget all made-up stories about sector-specific shocks. The only thing that matters is the distribution of price changes, its mean and its dispersion. Two months ago we wrote: “There is no reason to minimize what it is happening in the Euro area, it is terrible at this point“. Unfortunately, we were not wrong, especially because it seems that the monetary authority keeps underestimating the issue.
Until the distribution will stabilize and travel back, we will remain skeptical than a month (or two) of “encouraging” data will mean anything (see what happened in the UK..). Net of noise and measurement errors, there is nothing right now in the Euroarea data that says we are going back to the 2% target.
Figure 1. Kernel of HICP excluding food and energy items changes (%, a.r.)
Figure 2. Median HICP price increase
Evidence from our CI-C model
Our CI-C model estimates that net of Covid and idiosyncratic shocks, the common component across items in March was solid and in line with the previous months. Figure 3 shows the decomposition of the MoM of core HICP in the “common” component, the “idiosyncratic” component, and the “Covid” effect. The model estimates that in March the common component increased by 16bps, in line with the average of the previous 16 months or so. The Covid effect is estimated at 30bps, and the idiosyncratic shock is positive (3bps). Because the idiosyncratic shock was pretty small this month, it is fair to say that the overall core HICP reading this month (around 50bps MoM) is a good representation of where the EA inflation process is: we are facing a core inflation process centered around 5%-6% at annual rate. Part of this process might “fix itself” (although we are skeptical) because driven by negative supply-shocks. But the remaining part is not an easy fix, it is a persistent component driven by positive aggregate demand shocks.
Figure 3. Contributions to MoM changes of HICP excluding food and energy items
Note: the Figure shows the decomposition of the MoM percent changes of HICP prices excluding food and energy items. The contributions are estimated using our CI-C model, a 2-stage OLS-LASSO regression model. The “Covid” effect is identified with price variations outside the 10th-90th percentiles of each item pre-Covid price change distribution.
Figure 4. Estimated “Common” component: YoY, 3m/3m a.r. and 6m/6m a.r.
Note: the Figure shows the 3m/3m at annual rate (green line), the 6m/6m at annual rate (red line), and the YoY (blue line) of the “common component” estimated using our CI-C model.
Implications for the medium-term forecast of core HICP
The medium-term forecast of core HICP is little changed as the final release was close to the flash estimate. Our EA “main” Phillips curve model forecast is at 5.4% (Q4/Q4) in 2023 (average of four quarters at 5.5%), 4.0% in 2024, and 3.8% in 2025.
Note: the confidence intervals (C.I.) are calculated using the estimated parameters distributions.
First quarter of forecast: 2023:Q2.
Implications for the ECB staff and the Governing Council
Is a 5% terminal rate so impossible for the ECB? In the US, the Fed has signaled a terminal rate above 5%. Today, markets are pricing a 5% terminal rate for the BoE after a “disaster” CPI report. In the meanwhile, the doves in Frankfurth are arguing that with another 25bps hike the ECB will be done. Our view and estimates have not changed: disinflating is an hard task. If there is anything we have learned from history is that inflation “is a bad client to deal with”. Since we have started covering the Euroarea we have argued that 4% terminal rate for the ECB was a lowerbound estimate, considering that the information set of the models is not as large as the one of the econometrician. We have not changed our view (in fact, if anything the probability of being in a fiscal dominance world are higher now than 6 months ago). We continue to think that the ECB will need to reach at least 4% and that fiscal policy should help. Not only, but given what is happening in the US and the UK we cannot exclude even more than that. By saying so we are not calling for a 5% terminal rate in the EA; but we are glad to see that some large investment banks have recently raised their terminal rate calls for the ECB, and markets are slowly pricing in our long-standing view.