The Great Unanchoring of 2022?
The rise of Inflation Expectations
In my last post, I speculated on the possibility that inflation expectations could become “unanchored” as a result of the high monthly rates of inflation we saw over the second half of 2021, which are likely to continue for the first few months of 2022.
By “unanchored” I meant that the public decides that they no longer believe the Federal Reserve is committed to maintaining a long run average inflation rate of (about) 2 percent. I reprint here how I described it before (referring to the supply chain problems of 2021):
In time, I feel certain the effect of the supply chain shocks on the rate of inflation will abate. However, that does not mean I think the shocks are harmless. To me, the real danger is not the shocks themselves, but that they may force the public to lose faith in the Fed, leading to an unanchoring of expectations. If the public is not patient enough with the Fed, or the Fed tests that patience by not acting with enough force against inflation (by moving to close the output gap), then inflation expectations very well could shoot up in 2022. In that scenario, the rates of inflation that we saw in 2021 will become long-lasting.
This begs the question: How can we tell if those expectations become unanchored?
Fortunately, there are a few metrics provided out there, some survey-generated metrics, and the others “market”-generated metrics. So, let’s check out some of those metrics to see what exactly the public is expecting for 2022 (recall, I describe here what I mean by the “public”).
Surveys of Consumers (University of Michigan)
One of the most-oft reported measures of inflation expectations is from the University of Michigan’s “Surveys of Consumers.” This survey provides a measure of inflation expectations for the coming year as well as for the next five years—the expectations are reported monthly (you can find detailed information on the survey here). Here is a picture of inflation expectations from the Survey of Consumers going back to 2000 (downloaded from FRED):
The “headline” of the graphic is that inflation expectations have jumped to 4.8 percent as of December 2021, up from 3.0 percent as measured in January of 2021. (Note, the value for January of 2022 stands at 4.9 percent, which is not shown in the figure above, but reported from the survey source here.)
You can compare the most recent numbers to the historical averages, shown on the graph. I report the averages for before and after the Great Recession, to get a sense of what was “normal” for the past twenty years. And of course, our recent experience is not normal. (Though not unprecedented as seen with the spike at the beginning of the Great Recession in 2008.)
Okay, that’s one source of “inflation expectations” of the public. Another is from the Fed.
The Cleveland Fed’s Expected Inflation Measure
The Federal Reserve Bank of Cleveland publishes an “Expected Inflation” forecast for various time frames—for the next year, two-years, five years and ten years. The Cleveland Fed generates their measures of inflation expectations from a combination of financial data, inflation data and surveys. The current forecasts and additional documentation are found here on the Cleveland Fed’s website. Here I focus on their one-year forecast, shown here:
The “headline” from this measure of inflation expectations is emphasized on the Figure: 2.44 percent for the next 12 months, as measured in January of this year. That stands in contrast to what was expected a year ago: 1.45 percent reported in January of 2021.
Hence, inflation expectations as measured by the Cleveland Fed are clearly higher than a year ago. And given the historical averages shown on the graphic, recent expectations are obviously higher than the monthly average over the 2009 to 2019 period (right after the Great Recession up to the last “normal” year before Covid-19). However, current inflation expectations are similar to the average for the first decade of the 2000s.
What is interesting to me is that in spite of historically high rates of inflation in the last few months—including the just-released estimate for January 2022—the expectation for inflation is not that far off the historical mark. At least, that is the case according to the Cleveland Fed’s assessment.
The Michigan measure, of course, is more dramatic, 4.8 percent relative to the just-below-3 percent forecasts for the earlier historic eras as highlighted on the figure above.
So, between these two measures of inflation expectations, what conclusion should we draw? Are expectations becoming “unanchored”?
If we define “unanchored” as how I describe it above, then it appears the public—at least for the next year—expects the Fed will not bring inflation down to the 2 percent level that is considered their “target.” By Michigan’s measure, that is clearly the case; by the Cleveland Fed’s measure, the deviation is less substantial (only 0.44 percent above the target). However, the survey-generated numbers are not that far off from the Fed’s own 2022 forecast for the rate of inflation—2.6 percent (as discussed here).
Hence, by virtue of this evidence, perhaps we should not be too concerned as of yet for a “Great Expectations Unanchoring.”
To explore just a bit more, let’s take a quick look at a different source, this time culled from “markets”—that is, culled from the collective behavior of investors all over the world. The graph below shows a composite of rates on long-term U.S. government bonds, for bonds with a maturity greater than ten years. The interest rate series is published by the U.S. Treasury (here).
Why are long rates important for judging inflation expectations? The idea is as follows: as a saver (a.k.a. investor), if you are going to lock your money away in an asset for more than ten years, you should be compensated for future inflation. Otherwise inflation will erode the value of your savings as it sits there. If so, why bother investing in a long-term bond (when you could put your money in shorter term assets instead)?
Hence, investors “build” their inflation expectations into long term bond rates. If they change their opinion about that future inflation, one should see long rates rise as a consequence. Investors, in such a case, are demanding to be compensated for the rise in expected inflation.
Other factors could explain a rise in long-term rates, such as increase in the risk of default by the bond issuer. However, with bonds issued by the U.S. government that risk is very low, even at long horizons like ten years. Hence, long-term bond rates are considered a decent proxy for capturing long term expectations for inflation.
(For a wonkish deep dive on the subject, see here. For a quick exploration, check out Investopedia’s explanation. Note also, for the purposes of this post I am ignoring what is called the Yield Curve—the difference in average long rates from average short rates—I will cover that in future posts.)
Here’s a picture of the aforementioned long rate:
In the figure above, the composite long rate is shown back to the beginning of 2021 (though you can grab the data on the Treasury’s website back to 2000 if you wish). I have two “at quick glance” takeaways from this series:
Long rates have increased since early November of 2021, consistent with a rise of inflation expectations, and the other survey-based measures discussed above.
Long rates also increased notably in the first three months of 2021, before inflation really started to take off.
So, perhaps the recent uptick reflects a change in inflation expectations, but we cannot say for sure. Of course, these are quick observations, not meant to be a deep-dive on the drivers of long rates. Rather, think of the long rate as a “tell”, a give-away that things may be brewing when it comes to the inflation expectations of financial market participants. Hence, the recent uptick of the long rate is, at least, moving in the same direction as the survey-based measures of expectations cited above.
Finally, where does that leave us? Are we on the cusp of a “Great Unanchoring”? Maybe, maybe not. In the least, there is plenty of evidence of a “reset,” at least. Whether or not that reset leads to a big jump beyond, say, the 4.8 percent from the University of Michigan’s Survey of Consumers, remains to be seen. All we can do is continue to monitor the macroeconomic data as it evolves; a task which this blog intends to do.