The recent drop of breakeven rates (60-70bps on the 10y since April) has triggered questions about how the Fed (staff) might react and how much signal one should take from it. For this reason, we have updated our mixed-frequency Dynamic Factor Model (DFM) of “Common Inflation Expectations” (CIE) nowcast to include new readings from the University of Michigan survey of consumers and TIPS breakeven rates. This is our second (and last) update for 2022:Q3.
Keep in mind
We estimate that the Fed CIE -the synthetic index used by the Fed staff to gauge the behavior of inflation expectations- has ticked down in the current quarter, although the downward movement is small and contains little signal (see below for details). The level of the Fed CIE remains elevated and just a bit above the level consistent with the Fed target.
Results
According to our model (see Figure 1 below), the level of the Fed CIE is estimated at 2.18 (C.I. 2.16-2.22) in Q3, down a touch compared to last quarter. We estimate that the current level of inflation expectations is consistent with a reading of core PCE price inflation of 2.3 percent.
The downward tick in Q3 is driven by the downward movement in market-based measures of inflation compensation (breakeven rates) while survey-based measures of inflation expectations have remained -on net- stable.
How much signal should we take from breakeven rates?
The Fed CIE is correlated with breakeven rates by construction because it is a synthetic indicator that -among other variables- includes the breakeven rates as raw series. All else equal, movements of breakeven rates result in a similar movement of the Fed CIE.
Having said that, breakeven rates are measures of inflation compensation, not of inflation expectations. Therefore, one cannot directly infer markets expectations from breakeven rates but must estimate them using a model that accounts for liquidity and risk premia. It follows that at any point in time, breakeven rates (and the Fed CIE) can send different (or opposite) signals compared to the “true” markets inflation expectation component.
How can markets inflation expectations be estimated and what is the current evidence?
There are two publicly available models maintained by the Fed (system) staff that estimate markets inflation expectations:
- The Don, Walsh and Wei (or DWK) (2019) model maintained by the Fed Board.
- The Haubrich, Pennacchi and Ritchken (or HPR) (2015) model maintained by the Cleveland Fed.
For brevity, we do not discuss the technical details of these models. We just remind the reader that the DKW model uses TIPS to extract the inflation component, while the HPR model uses swaps. Figure 2 shows the latest estimates of both models. The main takeaway is that according to the models, markets inflation expectations have dropped below the Fed target during the pandemic, but they have increased since then and they are now above target (the DKW model suggests expectations are still trending higher while the HPR model suggests they are going sideways).
Implications for the Fed staff
The Fed staff has constructed the Fed CIE as a synthetic indicator of several series of inflation expectations/compensation. In our view and experience, the CIE offers guidance of the message that the Fed staff delivers to the FOMC about long-term expectations. As mentioned above, we estimate that the Fed CIE has ticked down in Q3 but the downward movement is small and driven by breakeven rates (the expectation component of which has probably not ticked down, as discussed).
For this reason, in our view the Fed staff will continue delivering the same message of the last few months: long-term inflation expectations appear roughly stable and anchored to a level consistent with a reading of core PCE price inflation a bit above target. On the one hand, this evidence continues to be good news for the FOMC because expectations appear anchored; on the other hand, however, the level of expectations continues to be the highest tolerable by the FOMC.
In conclusion: the evidence suggests the Fed cannot pivot now because there are too many upside risks. At least, certainly not because breakeven rates have fallen in recent months.
Figures
Figure 1. Published Fed-CIE and our model-based nowcast
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:Q2. The latest circle in the Figure shows our nowcast for 2022:Q3. 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 normalized at the beginning of estimation. Instead, the level of the common factor can be compared to its own history.
Figure 2. DKW (2019) model and HPR (2015) model estimates of long-term (10y) markets inflation expectations
Note: The figure shows the estimates model-based markets inflation expectations (10y horizon). The yellow line refers to the model of Kim, Walsh and Wei (or DWK) (2019) model maintained by the Fed Board. The brown line refers to the Haubrich, Pennacchi and Ritchken (2015) model maintained by the Cleveland Fed. For both models, the latest observation is September 2022.