HY INFLATION EXPECTATIONS - European Central Bank€¦ · time, so expectation of higher inflation...

INFLATION EXPECTATIONS – A POLICY TOOL?

Olivier Coibion (UT Austin and NBER)

Yuriy Gorodnichenko

(UC Berkeley and NBER)

Saten Kumar (Auckland University of Technology)

Mathieu Pedemonte

(UC Berkeley)

WHY INFLATION EXPECTATIONS?

Key variable for economic decisions: perceived real interest rate

𝑖 𝐸 𝜋

Conventional monetary policy:

• Anchor inflation expectations 𝐸 𝜋 • Increase/decrease nominal interest rate 𝑖

𝑖 𝐸 𝜋

Unconventional monetary policy

• Manipulate inflation expectations 𝐸 𝜋 • Nominal interest rate is at the zero lower bound (ZLB)

𝑖 𝐸 𝜋

Unconventional monetary policy

• Manipulate inflation expectations 𝐸 𝜋 • Nominal interest rate is at the zero lower bound (ZLB)

Mario Draghi (2015): “When inflation expectations go up with zero nominal rates, real rates go down. When real rates go down, investments and the economic activity improves. That’s the reasoning [of QE].”

STANDARD MECHANISMS Households consume more: when inflation expectations rise and nominal interest rates

are unchanged (ZLB), real interest rates are lower, so households should save less and spend more.

Firms invest more and hire more workers: when inflation expectations rise and nominal interest rates are unchanged (ZLB), real interest rates are lower so user cost of capital and labor are lower, inducing firms to raise their capital and employment.

Firms raise their prices: with sticky prices, inflation lowers firms’ relative price over time, so expectation of higher inflation induces them to raise prices more than they would otherwise.

Workers raise their wage demands: with sticky wages, inflation lowers the real wage over time, so expectations of higher inflation induce workers to raise wage demands, which should raise prices further.

SOME PRACTICAL QUESTIONS Whose expectations should central banks use for policymaking?

• Households and firms (and financial markets)

What influences household/firm inflation expectations when inflation is low?

• Salient prices of frequently-purchased, homogenous goods

Can central banks manipulate inflation expectations?

• Yes. Households and firms revise their beliefs in response to incoming information (but central banks often fail to reach households and firms)

Do economic agents react to changes in inflation expectations?

• Yes. They update their consumption/employment/pricing/etc. (but mechanisms remain unclear)

What are the challenges in using inflation expectations as a policy tool?

What are the challenges in using inflation expectations as a policy tool? • Measurement of inflation expectations (especially firms) • Breaking through the veil of inattention

WHOSE INFLATION EXPECTATIONS?

1980 1985 1990 1995 2000 2005 2010 2015

Asset Prices (Cleveland Fed)Professional forecasters (SPF)

1980 1985 1990 1995 2000 2005 2010 2015

Asset Prices (Cleveland Fed)Professional forecasters (SPF)Households (MSC)

Expectations are not interchangeable across agents

1980 1985 1990 1995 2000 2005 2010 2015

Asset Prices (Cleveland Fed)Professional forecasters (SPF)Households (MSC)Firms

WHAT FORCES INFLUENCE INFLATION EXPECTATIONS?

Predictors of inflation expectations

Perceptions/experience of inflation

Shopping

flatio

1990 1995 2000 2005 2010 2015

Households (MSC)Gasoline price (right axis)

Shopping

Policy

Perceptions/experience of inflation (strong)

Shopping (strong)

Media (intermediate)

Policy (weak)

CAN WE MANIPULATE INFLATION EXPECTATIONS?

Randomized controlled trials: provide subsets of firms or households with various bits of information and see how they update beliefs about inflation

Informational

treatment (inflation

target)

Elicit expectations (posteriors)

Elicit expectations

(priors)

CAN WE MANIPULATE INFLATION EXPECTATIONS?

Randomized controlled trials: provide subsets of firms or households with various bits of information and see how they update beliefs about inflation

Upshot:

• Households and firms appear to be Bayesian learners

Strength of priors Strength of signals

• Information effects are short-lived (less than 6 months)

Posterior beliefs

CAN WE MANIPULATE INFLATION EXPECTATIONS? Coibion, Gorodnichenko and Weber (2018)

Dep. var.: Revision of one-year-ahead inflation forecasts of U.S. consumers Treatment groups (coefficients are relative to control) Past inflation

-1.954*** (0.366)

Inflation target

-1.411*** (0.341)

FOMC inflation forecast

-2.004*** (0.384)

FOMC statement

-2.272*** (0.335)

USA Today coverage of FOMC statement

-0.950** (0.397)

-1.954*** (0.366)

Inflation target

-1.411*** (0.341)

-2.004*** (0.384)

FOMC statement

-2.272*** (0.335)

-0.950** (0.397)

-1.954*** (0.366)

Inflation target

-1.411*** (0.341)

-2.004*** (0.384)

FOMC statement

-2.272*** (0.335)

-0.950** (0.397)

HOW DO CONSUMERS AND FIRMS ACT ON THEIR INFLATION EXPECTATIONS?

Identification challenge: most of the time we see correlations

Causal estimates are scarce

• Natural experiments (fiscal policy)

• Randomized controlled trials

Causal estimates are scarce

• Natural experiments (fiscal policy)

• Randomized controlled trials

Informational treatment (inflation

target)

Elicit expectations (posteriors)

Elicit expectations (priors) and

Measure outcomes

(relative to plans)

Consumers: higher expected inflation → higher consumer spending

Firms: higher expected inflation → • New Zealand (2014-2016): higher employment and investment no effect on prices

• Italy (2012-2018) lower employment, lower investment plans higher prices

• France (1992-2016) higher prices

Conclusions: • Households/firms react to changes in inflation expectations

Conclusions: • Households/firms react to changes in inflation expectations • Manipulating inflation expectations can have a direct effect on prices

Conclusions: • Households/firms react to changes in inflation expectations • Manipulating inflation expectations can have a direct effect on prices • … but need more research on mechanisms and general equilibrium effects

CHALLENGES IN USING INFLATION EXPECTATIONS AS A POLICY TOOL?

Measurement of inflation expectations (especially firms)

We have reasonably good measures of inflation expectations for households

We have poor measures of inflation expectations for firms

• Small, non-representative samples Example: Industry Trends Survey (Confederation of British Industry)

• Qualitative questions about inflation (“same”, “up”, “down”) Example: Business Survey (European Commission)

• Priming of responses (e.g., restrict possible responses) Example: Business Outlook Survey (Bank of Canada)

• Questions about firm-specific outcomes rather than aggregate outcomes Example: Business Inflation Expectations (BIE; FRB Atlanta)

• Small, non-representative samples

• Qualitative questions about inflation (“same”, “up”, “down”)

• Questions about firm-specific outcomes rather than aggregate outcomes

• Priming of responses (e.g., restrict possible responses)

Surveys of firms’ inflation expectations are expensive but they are most useful!

Good surveys: Ukraine and Uruguay

Measurement of inflation expectations (especially firms) Breaking through the veil of inattention

BREAKING THROUGH THE VEIL OF INATTENTION Perceived inflation target in the U.S.

0 1 2 3 4 5 6 7 8 9 10+ DNK

FirmsHouseholds

BREAKING THROUGH THE VEIL OF INATTENTION

2011m1 2011m7 2012m1 2012m7 2013m1

January 2012: Fed annouces 2% IT

BREAKING THROUGH THE VEIL OF INATTENTION

Source: Deloitte Survey of Chief Financial Officers

2015h2 2016h1 2016h2 2017h1 2017h2 2018h1

GermanyFranceItalySpainFinland

2015h2 2016h1 2016h2 2017h1 2017h2 2018h1

NorwaySwedenPolandRussiaTurkey

Employment outlook; ECB announces QE tapering

BREAKING THROUGH THE VEIL OF INATTENTION Communication can work

• Public campaigns for other policies (healthcare, fiscal) work • Information treatment moves expectations

Simple messages are better • simple messages/facts are as effective as complex policy statements

Dep. var.: Revision of one-year-ahead inflation forecasts of U.S. consumers Treatment groups (coefficients are relative to control) FOMC inflation forecast

-2.004*** (0.384)

FOMC statement

-2.272*** (0.335)

Simple messages are better

Repeat the message • One-time announcements do not have long-lasting effects on expectations

of firms and households (information “depreciates” within months) • Need information campaigns

Repeat the message

Take the message directly to the target audience • Conventional media may be not good enough

Dep. var.: Revision of one-year-ahead inflation forecasts of U.S. consumers Treatment groups (coefficients are relative to control) FOMC statement

-2.272*** (0.335)

-0.950** (0.397)

Repeat the message

Take the message directly to the target audience • Conventional media may be not good enough → Advertising, social media

Dep. var.: Revision of one-year-ahead inflation forecasts of U.S. consumers Treatment groups (coefficients are relative to control) FOMC statement

-2.272*** (0.335)

-0.950** (0.397)

Repeat the message

• Potentially differentiate messages across audiences

Repeat the message

Currency bloc with booming “North” and lagging “South”

Repeat the message

One interest rate as a tool is not enough to address regional imbalances…

Repeat the message

One interest rate as a tool is not enough to address regional imbalances…

But one can have different messages for “North” to slow down and for “South” to accelerate.

Repeat the message

Target the message to the scenario • For example, if inflation is too low, emphasize the inflation target (rather

than actual inflation) to raise inflation expectations

CONCLUDING REMARKS Inflation expectations as a policy tool: large potential!

• Move consumption/employment/investment

• Directly influence prices

• Target specific areas, industries, or types of consumers

Are we ready to use it?

Are we ready to use it? Not yet…

• More research to study how inflation expectations translate into actions

• More high-quality surveys of firms’ inflation expectations

• New communication strategies to reach consumers and firms

HY INFLATION EXPECTATIONS - European Central Bank€¦ · time, so expectation of higher inflation...

Documents

RepulsiveGuidanceMoleculea(RGMa)Induces

Motor Learning induces plasticity in the resting brain ...eprints.chi.ac.uk/1784/3/Motor Learning induces plasticity in the... · Motor Learning induces plasticity in the resting

Inflation, Inflation Expectations, - FRASER Inflation Expectations, ... careers | site map ... policymakers not to treat inflation as a static concept,

Core Inflation and Trend Inflation Appendixmwatson/...trend_inflation_online_appendix_20… · Core Inflation and Trend Inflation − − Appendix − − June 2015 (Revised November

EVERYONE ECONOMICS It’s Time To Give Millennials A Raise - … · 2018-11-19 · 2 Generation Progress It’s Time To Give Millennials A Raise inflation.8 In SeaTac, WA, a successful

Intercellular communication induces glycolytic

Inflation Dynamics: The Case of Egypt - uni-muenchen.de€¦ · Demand pull inflation; 2. Cost push inflation and; 3. Inflation driven by inflation expectations. The relative importance

PUTIN’S PERILOUS PROLEM€¦ · resource development, restrict foreign investment, raise the cost of money, and generally increase inflation. 2014 Crimea & Ukraine Invasion Sanctions

Load & Inflation - Table B - Bridgestone & inflation load & inflation firestoneag.com English units of mEasurE tirE sizE inflation load limits at various cold inflation pr Essur s

Enriched environment induces angiogenesis

Prices and Inflation€¦ · 4.4 Food Inflation in the country has been extremely benign. Even globally, food inflation has been moderate. Food inflation Table 1: General inflation

Current Economic Environment - Inforum€¦ · attempts to raise inflation and spur growth that included policy interest rates near zero for seven years and “quantitative easing”

Inflation, types of inflation

Drivers of Inflation Compensation: Evidence from Inflation … · · 2017-01-03Drivers of Inflation Compensation: Evidence from Inflation Swaps in ... Drivers of Inflation Compensation:

Ch. 12: Inflation Types of inflation –Demand-pull inflation –Cost-push inflation Effects of inflation SR and LR long-run relationships between –inflation

Inflation, Inflation, Inflation Graeme Troy FFA April 2010

Inflation And Types of Inflation

THE INFLATION TARGETING DEBATE - Columbia · PDF filethat experienced high inflation adopted inflation targeting in order to get inflation under control and inflation targeting

Inflation - MPA :: MPA Startseite · Plan of the lectures: Inflation: a general outlook Basic inflationary models (new inflation, chaotic inflation, hybrid inflation) Creation of

Axoplasm Enriched in a Protein Mobilized by Nerve Injury … · · 2003-09-09Axoplasm Enriched in a Protein Mobilized by Nerve Injury Induces Memory ... raise interesting general