![](https://csdnimg.cn/release/download_crawler_static/88095840/bg10.jpg)
16
Economic Research
US: Not all inflation expectations
are created equal
May 3, 2019
JPMorgan Chase Bank NA
Michael Feroli (1-212) 834-5523
michael.e.feroli@jpmorgan.com
Economic Research Note
US: Not all inflation
expectations are created equal
Preventing a drift lower in inflation expectations is a
key goal of the upcoming Fed strategy review
If the outcome of that review only bolstered financial
market expectations, the benefits would be limited
A J.P. Morgan survey indicates the response of financial
market inflation expectations may be limited
Nonfinancial business expectations get little attention,
but probably warrant a closer following
Inflation expectations have taken pride of place in the Fed’s
review of its monetary policy strategies. Given the attention it
gets from Fed officials, one might even conclude that inflation
expectations are more important than actual inflation. Coun-
terintuitively, we see the emphasis on inflation expectations
reflecting concerns about employment, and about the Fed’s
ability to supply enough stimulus in the next downturn to pre-
vent another crushing loss of jobs. To see how the labor mar-
ket and inflation expectations are related, consider two states
of the world: one in which you expect inflation to be 1%, an-
other in which you expect inflation to be 3%. The next time
the Fed has to cut rates to near zero, in which of the two states
of the world would you be more likely to, say, take out a
mortgage to buy a house or vacation property?
Hopefully you said the 3% inflation expectations environ-
ment. If everyone else is equally financially savvy, then ag-
gregate demand would be stronger in the next zero lower
bound episode—and presumably the loss of employment
would be less—than in the 1% inflation expectations world.
The latter would be closer to Japanization. This should also
underscore why “admitting reality” and lowering the inflation
objective is contrary to basic economic reasoning: it would
leave real rates higher in the next zero-bound episode. The
importance of the state of inflation expectations heading into
the next downturn puts a premium on the Fed making sure
those expectations are well supported now, while the sun is
still shining.
That’s not only easier said than done, it’s not even clear when
it’s done. As is well known, inflation expectations are very
poorly measured. Nor is there clear consensus on whose ex-
pectations matter. Financial market participants’ inflation
expectations may be the most sensitive to Fed communica-
tions. This raises a risk that if the Fed does try to raise infla-
tion expectations, nominal interest rates could move higher
even if non-financial businesses and households didn’t change
their inflation expectations in response to Fed communica-
tions. This would increase their perceived real interest rates.
We make use of a new data set to demonstrate why we don’t
think this concern should prevent the Fed from moving to-
ward an average inflation targeting framework. Another issue
related to the issue of the segmentation of inflation expecta-
tions is whose expectations should matter. For understanding
where inflation is headed, common sense suggests that the
expectations of price-setters, i.e., businesses, should receive
some prominence. We present some evidence consistent with
this conjecture. Regrettably, these expectations have received
less attention from both policymakers and survey designers.
You talking to me?
If you are reading this you are likely (a) employed in the fi-
nancial services industry and (b) more attuned to Fed policy
than the average American. This could create a problem. Con-
sider a simple case where the Fed announces that it is raising
its inflation target. Longer-dated nominal interest rates might
rise, as Fed-sensitive market participants demand extra infla-
tion compensation to hold nominal assets. However, for
households and non-financial businesses who don’t hang on
the FOMC’s every word, some of this rise in nominal interest
rates might be perceived as an increase in real interest rates,
as they are unlikely to factor in higher inflation expectations
simply because the Fed said so.
To the best of our knowledge, this topic was first studied in a
paper presented by three senior Fed researchers at the 2013
Jacques Polak conference. They found that in such a world of
segmented expectations, the announcement of a new, 1%-pt
higher, inflation target would indeed depress output and slow
any recovery. Even so, such a policy might be desirable be-
cause the eventual recovery to a higher inflation rate and
higher inflation expectations would reduce the frequency with
which monetary policy would be constrained by the zero
bound in the future.
In a recent working paper, Fed chair Bernanke and his co-
authors also examine the impact of financial markets knowing
more about the Fed’s reaction function than ordinary folks.
They find that for a variety of policy strategies that are de-
signed to bolster the credibility of the Fed’s inflation objec-
tive—such as price-level targeting or temporary price-level
targeting—the segmentation of knowledge about the Fed’s
reaction function reduces, but does not eliminate, the benefit
of adopting such strategies.
Some more insight into these issues can be gleaned from our
latest semiannual survey of inflation expectations (link here).
This survey asks the inflation expectations of money manag-
ers, most of whom manage over $5 billion in assets and many
of whom are active in inflation-linked markets. These are pre-