As expected, the Federal Open Market Committee (FOMC) announced no changes to its administered rates following its April meeting, and Federal Reserve Chair Jerome Powell did not provide new information about the Fed’s bond-buying programs. The FOMC reaffirmed it wants to observe substantial progress toward its dual mandate on employment and price stability, including assurances that the worst of the pandemic is behind us, before scaling back accommodation.
While this wasn’t a new message from Powell, it’s nonetheless notable that the Fed is so focused on backward-looking data instead of forward-looking projections, since the outlook has improved so substantially and monetary policy works with long and variable lags. However, we think this new focus is a necessary condition for the Fed’s ultimate success, primarily due to the importance of anchoring long-term inflation expectations.
The end of ‘copacetic coincidence’
Since the formal introduction of the Fed’s 2% inflation target in 2012, median FOMC estimates of the neutral real interest rate (r*) have declined substantially from 2.25% in January 2012 to 0.5% today. The implications of this decline are profound: The FOMC now has less room to accommodate adverse economic shocks by cutting its benchmark interest rate. And this, in turn, increases the likelihood that the Fed will be constrained by the zero lower bound in the future, and elevates the risk that the U.S. economy will face more frequent periods of disinflation or outright deflation that over time depress inflation expectations. These risks are especially acute under a strictly forward-looking inflation-targeting strategy – the Fed’s previous approach – since 2% acts more like a ceiling than a long-term average. In this case, the resulting drift lower in inflation expectations further tightens the constraints on the central bank.
Fed Vice Chair Richard Clarida discussed this in an April 14 speech, when he declared the end of “copacetic coincidence.” In a paper he wrote in 2003, Clarida found that absent a zero-lower-bound constraint, a monetary policy that targets actual inflation should deliver long-run inflation expectations well-anchored at the target “for free.” However, Clarida noted, in a low r* world, this copacetic coincidence no longer holds.
This explains the importance to the Fed’s broader monetary policy strategy of longer-term inflation expectations. It also explains why the Fed shifted last year to a flexible average inflation-targeting approach, which aims for above-target inflation after periods when inflation runs persistently low. However, important questions related to inflation expectations remain: How anchored are long-term inflation expectations? And what are they currently implying for realized inflation, and the Fed’s ability to achieve its target? In response to these questions, we would make a few points:
- Inflation expectations appear to adapt slowly to trends in realized inflation.
- As a result of the period of persistently below-target U.S. inflation following the global financial crisis, inflation expectations appear to have re-anchored under the Fed’s longer-run 2% target.
- If the Fed wants to sustainably reach its target, it must raise inflation expectations.
- Raising inflation expectations likely will require a period of above-target inflation.
- Additional fiscal stimulus may be exactly what it takes to jump start the period of above-target inflation.
Recently Vice Chair Clarida stated that, in his view, once normalization begins, the pace of Fed rate hikes will depend critically on the extent to which inflation expectations are rising. While some have interpreted this to mean that the Fed will view any rise in inflation expectations as a worrisome indicator of broader inflation risks, our analysis suggests it’s a necessary condition for the Fed to achieve its inflation target. Indeed, the Fed would likely prefer to see inflation expectations that slowly move back toward the 2% target.
How anchored are long-term inflation expectations?
Since inflation expectations can’t be directly observed, various household and business surveys devote numerous questions to the subject. While these surveys are informative, their potential biases make them difficult to interpret in the context of the Fed’s inflation goal. For example, over the last 20 years, the University of Michigan’s five- to 10-year average inflation expectation measure has varied from 2.5% to 3% – well above the Fed’s 2% long-term inflation target. However, if instead of looking at individual survey results, we focus on co-movements among the observations of multiple surveys, it appears U.S. inflation expectations have drifted lower since the financial crisis – a potentially worrisome development in the context of monetary policy.
As a result of these issues with interpreting survey data, Fed staff have created what they call the Common Inflation Expectations (CIE) Index, which uses statistical techniques to distill the information from the various surveys into one indicator. At PIMCO, we recreated this index, but instead of projecting it on the Survey of Professional Forecasters measure, which has systemically over-projected actual inflation since the 2000s, we projected it on realized core CPI (consumer price index) inflation (see chart, and for more on the Fed’s methodology, see FEDS Notes, Index of Common Inflation Expectations).
Comparing our CIE with core CPI inflation over time suggests that although inflation expectations historically have been relatively stable, in 2014 they moved down to around 1.75% and are now more consistent with inflation below the Fed’s target (see chart). Recall that the Fed targets 2% PCE inflation, or personal consumption expenditures, a measure that tends to run slightly below CPI.
Market-based measures tell a similar story
Important signals on inflation expectations can also be extracted from the market for U.S. Treasury Inflation-Protected Securities (TIPS). Unlike the survey measures, market-implied measures can embed biases associated with TIPS market illiquidity and the value investors place on hedging portfolios against inflationary environments.
To deal with these biases, Fed staff have also devised statistical procedures to extract the inflation expectations by filtering out noise from other liquidity and inflation risk premium factors. Based on this model, inflation expectations derived from TIPS markets have exhibited a downward trend over the last 20 years (see FEDS Notes, Tips from TIPS: Update of Discussions). Meanwhile, we believe the more recent rise in the five-year TIPS breakeven rate, five years forward has been driven by changes in inflation risk and liquidity premia more so than a reacceleration in inflation expectations.
Takeaways, risks, and implications
A range of indicators derived from surveys and markets suggests that U.S. inflation expectations may well be anchored under the Fed’s goal, and the Fed likely needs a period of above-target inflation to re-anchor inflation expectations at the long-run target. In our view, this is why the Fed is more focused on actual observed progress rather than forward-looking forecasts when assessing the appropriate timing of policy normalization.
In this context, fiscal stimulus that is larger than estimate for the current output gap (and that generates a period of above-target inflation) may be exactly what is needed to “jump start” the process by which inflation expectations slowly move back to target. But this process is still likely to take time. And as a result, Fed policymakers may welcome periods where inflation is modestly above target, and we wouldn’t expect them to act quickly to offset it.
However, engineering higher inflation expectations doesn’t come without practical risks, and the Fed may have to walk a fine line to ensure inflation expectations don’t overshoot the 2% target. In a worst-case scenario, above-target inflation causes an adverse feedback loop, whereby higher inflation expectations beget higher inflation, which, in turn begets even higher inflation expectations, and so on. It’s also possible that the Fed misinterprets the signals coming from the surveys, and inflation that is initially viewed as temporary turns out to be more persistent.
While our base case expectation (as we discuss in a recent Viewpoint) is that realized inflation remains relatively benign, the uncertainty around this expectation is high. Nonetheless, based on all of our empirical observations, we think the Fed is right to view the possible medium-term benefits of their new strategy as worth the risks of an unwanted spike in inflation.
Visit PIMCO’s inflation page for further insights into the inflation outlook and investment implications.
Tiffany Wilding is an economist focused on North America and a regular contributor to the PIMCO Blog.