We have updated our “Pre May 2022 Meeting Package” following the data releases of the last few days (Q1 GDP, March PCE prices, Q1 ECI, TIPS breakeven rates, and final April Michigan). Please, find below the updated models outputs.
To keep in mind
The information received in the last few days have not materially changed any aspect of the inflation outlook.
The incoming data on PCE prices have been close to our expectations, just a touch lower. However, the downward surprise is driven by “non-market” prices (while market-based prices continued to be very strong). For this reason, in our view the Fed staff has taken no signal and offset most of the surprise. All told, in our view the information of the last few days will make no difference for the upcoming FOMC.
March PCE prices
The incoming data on core PCE prices have been close to our expectations. In our pre May FOMC meeting package, we estimated the YoY of core PCE price inflation at 5.2% in March, in line with published data.
(Technical note: the March personal income report showed downward revisions in January and February, while March was marginally stronger than expected. For this reason, the YoY of core PCE prices in March came in as expected but the QoQ (in Q1) was a bit weaker than expected due to the quarterly arithmetics).
Takeaways from the monthly distributions:
- In March core PCE prices expanded 29bps (MoM), in line with (revised) February and below the average of the previous 4 months.
- In March, the percentiles of the distributions moved down in the left shoulder/tail but upward in the right shoulder/tail (for instance, the 5th pct decreased from -21.8% (MoM a.r.) to -24.5% (MoM a.r.) but the 95th pct increased from 44.9% (MoM a.r.) to 55.5% (MoM a.r.)). In other words, in March price dispersion as measured by the standard deviation of the distribution increased notably (which is a bad news for the Fed).
- The median of the distribution moved down from 3.7% (MoM a.r.) to 2.9% (MoM a.r.) – see figure below. In any case, the median remains elevated.
Note: the Figure shows the median (%, a.r.) of the distribution of PCE price changes excluding food and energy items (left panel) and the MA(12) of the median (right panel).
The role of non-market prices
Note: the Figure shows 3m/3m (a.r.), the 6m/6m (a.r.), the YoY and the 2Yo2Y (a.r.) of PCE prices market-based excluding food and energy items.
Takeaways from the CI-C model:
Consistent with the evidence from the monthly distributions, in March our CI-C model estimates that the common component expanded 19bps, in line with the previous months. The model suggests that, net of Covid-effects and idiosyncratic shocks, the most persistent part of inflation continues to be strong. The 3m/3m (a.r.) of the common component is estimated at 2.44% in March, 1 tenth lower than in February but in line with the readings of the last 10 months – see figures below.
Note: the Figure shows the decomposition of the MoM percent changes of PCE 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.
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.
“Main” Phillips curve model
As explained above, the 2022:Q1 (QoQ) of core PCE prices is a touch lower than expected before the March personal income report. The downward surprise is small but it did trigger a marginal revision to the model forecast. The model now expects core PCE price inflation (Q4/Q4) at 4.1% in 2022, 3.0% in 2023, and 2.7% in 2024. This forecast is essentially identical to the March FOMC round.
A. C.I. calculated from parameters distributions
Note: the confidence intervals (C.I.) are estimated using 100,000 draws from the estimated parameters distributions to simulate the path of core PCE price inflation going forward.
First quarter of forecast: 2022:Q2.
B. C.I. calculated using quasi-out-of-sample methods
Note: the confidence intervals (C.I.) are estimated using quasi-out-of-sample methods (estimate the model over a sub-sample, forecast, and calculate the root mean squared forecast errors).
First quarter of forecast: 2022:Q2.
Common Wage model
The data received on nominal wage growth in the last week have been strong. In 2022:Q1, the Employment Cost Index (ECI) increased 1.4% QoQ (and 4.5% YoY). Following the ECI report, our mixed-frequency dynamic factor model has reached 4.8% in Q1, 2 tenths higher than the day before the release. The model suggests that nominal wage growth is now higher than the 1990s and it is comparable to the 1980s. In our view, at these levels it appears possible that the US economy is already in some initial form of wage-price spiral.
Note: the Figure shows four measures of wages and total compensation, as well as the estimated common factor from our Dynamic Factor Model. ECI stands for “Employment Cost Index, Total Compensation, Private Industry, All Workers, SA”. AHE stands for “Average Hourly Earnings, All Employees, Total Private, SA”. CPH stands for “Nonfarm Business, Hourly Compensation, SA”. AFWT stands for “Atlanta Fed Measure of Median Nominal Wage Growth, Unweighted Overall, 3 Month Moving Average”. The four measures of wages and total compensation are shown at quarterly frequency in YoY growth rates. The common factor is specified as an autoregressive process of order 4 and it is projected onto the ECI (the measure with the lowest standard deviation). By construction, the absolute level of the estimated common factor cannot be interpreted because the four series are normalized at the beginning of estimation given their different means. Instead, the level of the common factor can be compared to its own history. Latest observation refers to 2022:Q1.
Fed CIE
The information received on TIPS breakeven rates and (final) April Michigan on long-term consumers expectations have not changed our nowcast of the Fed CIE.
According to our mixed-frequency dynamic factor model, the level of the Fed CIE is broadly consistent with a reading of core PCE price inflation of 2.1-2.2 percent.
Note: the Figure shows the published Fed Common Inflation Expectations (CIE) index (the thick blue bar), and our Dynamic Factor Model (DFM) nowcasts (the black circles). Both series are displayed at quarterly frequency. The latest published quarter of the Fed-CIE is 2022:Q1. The latest circle in the Figure shows our nowcast for 2022:Q2. The dashed-blue lines are estimated confidence intervals. The Fed-CIE model is a DFM that includes 21 measures of inflation expectations. The estimated common factor (which is specified as an autoregressive process of order 4) is projected on a chosen measure of long-term inflation expectations (the SPF PCE 10 years). By construction, the level of the common factor cannot be interpreted because the underlying series are demeaned (in fact, normalized) at the beginning of estimation. Instead, the level of the common factor can be compared to its own history. Our model is run on long-term measures of inflation expectations (8 in total) included in the Fed-CIE because these are the series driving the results.