FLEXIBLE INFLATION TARGETING AND ALTERNATIVE MONETARY

POLICY TARGETS – WHAT DOES RESEARCH TELL US?

ØISTEIN RØISLAND

Disclaimer

Views and conclusions expressed in this presentation are my

own should not be taken to represent the views of Norges Bank

Theory: Welfare-maximizing monetary policy can be implemented

through FIT– Woodford

Caveat 1: Welfare-maximizing monetary policy imply stabilizing a large

number of variables

Caveat 2: FIT might require considerable flexibility to implement

welfare-optimizing policy– “Divine coincidence” only in very simple, unrealistic models

Flexible Inflation targeting (FIT)

Price-level targeting– Or a less extreme variant: average inflation targeting

Nominal GDP targeting

– Level

– Growth

Alternatives to FIT

Price-level targeting

Inflation Price level

Price-level targeting

Resembles optimal policy under commitment

– History-dependence

Easier to escape a liquidity trap

– Expectations of expansionary future monetary policy

Less uncertainty about the future price level– Smaller unexpected changes in real value of debt

Advantages

Price-level targetingDisadvantages

Can lead to large fluctuations output (and inflation) if– expectations are not forward-looking, or

– price-level target not credible

More difficult to understand for the public

Has not (really) been tested in practice

Nominal GDP targetingAdvantages

Hard-wires a balance between price stability and output stability – PY = target

Resembles optimal policy under commitment– History-dependence

Easier to escape a liquidity trap

Reduces financial risk– Debt contracts based on fixed nominal repayments

Good properties for small open economies hit by ToT shocks– Difference between GDP deflator and CPI

Correlated with house prices and credit– Less need for “leaning against the wind”

Nominal GDP targetingDisadvantages

More difficult than IT to explain to the public

Large revisions in NGDP data

NGDP-level targeting has same problems as price-level

targeting if expectations are non-rational or limited credibility

NGDP and CPI tell different stories about nominal stability

10Sources: Statistics Norway and Norges Bank

0

2

4

6

8

10

0

1

2

3

4

5

6

1999 2002 2005 2008 2011 2014

Core CPI (l.h.s)

NGDP mainland (r.h.s.)

Annual change. Percent

Positively correlated with terms of trade shocks

11

Annual change. Percent

-6

-4

-2

0

2

4

6

0

2

4

6

8

10

12

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

NGDP mainland (l.h.s.) ToT mainland (r.h.s.)

Source: Statistics Norway

…and credit growth

12

Four-quarter change. Percent

-4

0

4

8

12

16

-4

0

4

8

12

16

1990 1993 1996 1999 2002 2005 2008 2011 2014

NGDP mainland

Domestic gross debt held by the public

Source: Statistics Norway

Difficult to beat flexible inflation targeting

Trade-off between flexibility and credibility/accountability

Research question: Could an alternative target (e.g. NGDP) provide sufficient real

and financial stability without requiring too much flexibility?

ReFIT research

Welfare loss under FIT and NGDP targeting for various degrees of flexibility

Brubakk and Røisland (2017)

Welfare

loss • NEMO

• Welfare depends on stability

in CPI inflation, output gap

and house prices (small

weight) (+ interest rate smoothing)

• Foreign and domestic shocks

St.dev. of target variable

Welfare loss under FIT and NGDP targeting for various degrees of flexibility

Bergholt (2017)

Welfare

loss

St.dev. of target variable(percentage points)

FIT

NGDP

• Three-sector model• Commodity (oil)

• Manufacturing

• Service

• Welfare depends on stability in CPI

inflation and output gap (+ interest rate smoothing)

• International shocks only

Conclusions

Welfare-optimal monetary policy can in principle be implemented through FIT– But needs to be flexible

Trade-off between flexibility and credibility/accountability

Price-level targeting and NGDP targeting have some nice properties in theory– But has some practical problems

Relatively strict NGDP targeting could come quite close to «optimal» FIT for

Norway– Correlated with house prices and credit: less need for “leaning”

– Implies appropriate monetary policy responses to ToT shocks

– NGDP as a cross-check on nominal stability?

FLEXIBLE INFLATION TARGETING AND ALTERNATIVE MONETARY

POLICY TARGETS – WHAT DOES RESEARCH TELL US?

ØISTEIN RØISLAND

NORGES BANK

Extra

NGDP and house prices

19

-2

0

2

4

6

8

10

12

14

16

-2

0

2

4

6

8

10

12

14

16

1999 2002 2005 2008 2011 2014

Nominal GDP growth

Houce price growth

Sources: Statistics Norway and Norges Bank

Price-level targeting

No weight on stabilizing house prices

Equal weight on inflation and house prices

Three-sector model (Bergholt)