National Bank of Belgium 12-13 October, 2006

Inflation persistence and price setting behaviour in the euro area

Frank SmetsEuropean Central Bank

National Bank of Belgium12-13 October, 2006

Introduction• Aim of this presentation

– Summary of current knowledge on inflation persistence and price setting behaviour in the euro area

– Based on the main findings of a collaborative Eurosystem research project: the Inflation Persistence Network (IPN)

• Objectives of the IPN:

– Investigate the degree of inflation persistence, defined as “The tendency of inflation to converge slowly towards its long-run value following shocks”;

– Study the patterns and determinants of price setting behavior, which underlie the dynamics of inflation in the euro area and its member countries

Introduction - IPN

• Collaborative research network of all National Central Banks of the Eurosystem and the ECB:

– Main activities from 2003 till 2005;

– Over 60 papers, of which more than 30 released as NCB and ECB Working Papers;

– For a comprehensive list, see http://www.ecb.int/home/html/researcher_ipn.en.html

http://www.ecb.int/home/html/researcher_ipn.en.html

http://www.ecb.int/home/html/researcher_ipn.en.html

ESLU NL PTDE GR IE ITFI FRAT BE EA

Micro-CPIData

Micro-PPIData

Firm-LevelSurveys

StructuralModels ofInflation

Dynamics

Macro TimeSeries

Synthesis of Results

Introduction – the datasets of the IPN

Introduction – Other IPN reading• Dhyne, Álvarez, Le Bihan, Veronese, Dias, Hoffmann, Jonker,

Lünnemann, Rumler, Vilmunen (2006), “Price changes in the euro area and the United States: Some facts from individual consumer price data”, Journal of Economic Perspectives, 20:2, Spring 2006.

• Fabiani, Loupias, Martins and Sabbatini (2006), “Pricing decisions in the euro area: How firms set prices and why?”, forthcoming Oxford University Press

• Fabiani, Druant, Hernando, Kwapil, Landau, Loupias, Martins, Mathä, Sabbatini, Stahl and Stokman (2006), “What firms’ surveys tell us about price-setting in the euro area”, International Journal of Central Banking

• Vermeulen, Dias, Dossche, Gautier, Hernando, Sabbatini, Stahl (2006), “Price setting in the euro area: Some stylised facts from individual producer price data and producer surveys”, forthcoming ECB Working Paper.

• Journal of European Economic Association, May 2006, pp.562-593

• Why study inflation persistence (IP)?

• Definition/measurement of IP

• Macro-evidence on the sources of IP

• Micro-evidence on price setting

• Policy conclusions

Overview

Inflation persistence – Why important?• For monetary policy

– A high degree of infIation persistence implies:

• a larger sacrifice ratio (higher output costs to stabilise inflation)

• the possibility that inflation expectations become unanchored.

– Understanding the sources of inflation persistence:

• Allows to build better macro-economic models and enables better forecasting of inflation

• Facilitates understanding the proper response of monetary policy to inflation shocks.

• For structural policies

– Understanding the rigidities underlying inflation persistence is important for guiding structural reforms.






Overview

Definition of inflation persistence

• The tendency of inflation to converge slowly towards its long-run value following shocks

• The long-run value is implied by the monetary policy regime

• The definition refers to the persistence of the deviation from the long-run value following a shock

• Making the definition operational

– Reduced-form inflation model (e.g. with one lag)

– Inflation persistence (or the speed of inflation convergence to the long-run value) depends on

ttt 1)1(

Measurement of inflation persistence

• Higher , higher degree of inflation persistence

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

quarters

0.9

0.7

Response of inflation following a 1percentage point shock


• Estimates of depend on assumptions regarding shifts in the mean of inflation,

-2

0

2

4

6

8

10

12

14

16

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

HICP inflation

Euro area


• Accounting for breaks in the mean reduces the estimated degree of inflation persistence

– Under the current regime, inflation persistence is moderate;

– there is some evidence that persistence has fallen since the mid 1990s;

– but a significant degree of uncertainty remains.

1.02-1.04 0.60-0.900.96 0.900.84 0.800.93 0.400.74 0.40

Masuch (2005) 0.65

O'Reilly & Whelan (2004)

Dossche & Everaert (2004)

Without breaks

Note: Parameters in bold indicate that it is possible to reject statistically that =1.

Robalo Marques (2004)Altissimo, Mojon & Zaffaroni (2004)Batini (2002)

Robalo Marques (2004)

Lünnemann & Mathä (2004)

Gadzinski & Orlandi (2004)

With breaks

Gadzinski & Orlandi (2004)O'Reilly & Whelan (2004)

Estimates of rho in the euro area






Overview

Sources of persistence – NK Phillips curve

• Three main sources:

– Extrinsic persistence:

• Inflation inherits the persistence of its approximate determinants such as the real marginal cost due to, for example, wage rigidities. A key factor is the degree of price stickiness:

– Intrinsic persistence:

• Temporary shocks lead to persistent inflation responses due to e.g. indexation or rule-of-thumb behaviour

– Expectations-driven persistence:

• Incomplete information and learning may lead to persistence coming from inflation expectations

ttttttt upmcpEpp )()1( 11

)( tt pmc

1 tp

1)1( tt pE

• Evidence based on estimation of the NKPC matches the reduced-form evidence

• Forward-looking behaviour is often dominant …

• e.g. Rumler, Jondeau & LeBihan, Gali, Gertler & Lopez-Salido

… but a high degree of uncertainty• e.g. Whelan

• Some evidence of higher degree of price stickiness (or Iower responsiveness of inflation to marginal cost) in the euro area (relative to the US)

• e.g. Gali, Gertler and Lopez-Salido

Sources of persistence: evidence

• Estimates of the importance of the backward-looking component or the degree of intrinsic inflation persistence fall– in stable monetary policy regimes; – with explicit measures of inflation

expectations and allowing for learning

• Inflation expectations do not appear to be a source of inflation persistence in the euro area, because they are well-anchored to the inflation objective.



0

1

2

3

4

5

6

HICP inflation and inflation expectations in the euro area

Annual HICP inflation

One-year ahead inflation expectations

Aggregate inflation dynamics: conclusions• Intrinsic inflation persistence low under the

current regime

• Well-anchored inflation expectations: source of “good” inflation persistence.

• Hence, inflation persistence mostly extrinsic or driven by the persistence of determinants

• But uncertainty due to large error bands, differences across studies

• The importance of extrinsic persistence suggests having a closer look at price setting at the micro level …






Overview

Patterns of price adjustments (1)• Prices change infrequently and less often than in the US:

Statistics Euro area US

CPI Frequency 15.1 24.8

Average duration (months) 13.0 6.7

Median duration (months) 10.6 4.6

PPI Frequency 20.0 n.a.

Surveys Frequency 15.9 20.8

Average duration (months ) 10.8 8.3

New Keynesian Phillips Curve Average duration (months) 13.5-19.2 7.2-8.4

Patterns of price adjustments (2)• Large heterogeneity across and within sectors, more pronounced than across countries

• Consumer prices

• Most frequent price changes for unprocessed food and energy, least for services

Euro Area 28.3 13.7 78.0 9.2 5.6 15.8United States 47.7 27.1 74.1 22.4 15.0 24.8

Non-energy industrial

Services TotalUnpro-

cessed foodProcessed

foodEnergy

(oil


Belgium 31.5 19.1 81.6 5.9 3.0 17.6Germany 25.2 8.9 91.4 5.4 4.3 13.5Spain 50.9 17.7 n.a. 6.1 4.6 13.3

France 24.7 20.3 76.9 18.0 7.4 20.9Italy 19.3 9.4 61.6 5.8 4.6 10.0

Luxembourg 54.6 10.5 73.9 14.5 4.8 23.0The Netherlands 30.8 17.3 72.6 14.2 7.9 16.2Austria 55.3 24.5 15.9 14.3 13.6 21.1

Portugal 37.5 15.5 72.3 8.4 7.1 15.4Finland 52.7 12.8 89.3 18.1 11.6 20.3

Euro Area 28.3 13.7 78.0 9.2 5.6 15.1

United States 47.7 27.1 74.1 22.4 15.0 24.8

Services TotalEnergy

(oil products)

CountryUnpro-

cessed foodProcessed

food

Non-energy industrial

goods


• Producer prices

• Most frequent price changes for energy, least for durables and capital goods

FoodNon

durable Non food

Durable products

Intermediate products

EnergyCapital goods

Belgium 0.20 0.11 0.14 0.28 0.50 0.13

France 0.32 0.10 0.13 0.23 0.66 0.12

Germany 0.26 0.14 0.10 0.23 0.94 0.10

Italy 0.27 0.10 0.07 0.18 na 0.05

Portugal 0.21 0.05 0.18 0.12 0.66 na

Spain 0.25 0.10 0.13 0.29 0.39 0.09

Euro area 0.27 0.11 0.10 0.22 0.72 0.09

Patterns of price adjustments (3)• Price changes are large relative to the

prevailing inflation rate

• Suggests that large firm-specific shocks are important;

• Frequency and size of price changes is higher at the consumer level than the producer level.

Average increase

Average decrease

CPI 8% 10%PPI 5% 5%

Patterns of price adjustments (4)

• Price decreases are common, except for services

• Similar for producer prices

Share of price increases 54 54 54 57 80 58

Size of price increases 15 7 3 9 7 8

Size of price decreases 16 8 2 11 9 10

Tot

al

Unp

roce

ssed

fo

od

Pro

cess

ed fo

od

Ene

rgy

(oil

prod

ucts

)

Non

-ene

rgy

indu

stria

l goo

ds

Ser

vice

s

Two possible reasons:1. Costs change infrequently2. Prices are sticky, even if the

underlying fundamentals change.

Why do prices change infrequently?

• Firm surveys confirm that:– Mark-up pricing is the dominant pricing strategy

• 54% in the euro area

– Moreover, relatively stable costs and cost-based pricing is reported as being an important reason (3rd in rank) for why prices change infrequently.

• The latter could also explain why the frequency of price changes is less than of price reviews:– The share of firms changing their prices less than

quarterly (max 3 times a year) is 86%, compared to 57% of firms reviewing their prices with the same frequency.

Why do prices change infrequently (1)?

Why do prices change infrequently? (1)• This is confirmed by cross-sectional analysis which relates price stickiness to the variability of input prices:– Higher labour share implies lower frequency

-.2

-.1

0.1

.2.3

Ad

just

ed

fre

quen

cy

-.2 -.1 0 .1 .2Adjusted labour share

coef = -.67961959, (robust) se = .13749755, t = -4.94

Data demeaned and adjusted for remaining regressors

Frequency of price change and labour share

Spanish PPI data

Why do prices change infrequently? (1)

• Similarly:– Higher share of raw material inputs implies

higher frequency;– Prices of products at low stages of

production change more frequently;

– VAT increases affect price adjustment frequencies

Why do prices change infrequently? (2)… but it is not the only story

• Survey evidence suggests that the most important reasons for firms not to change their prices even when the underlying determinants have changed are:

– Implicit and explicit contracts with long-term customers (1st and 2nd rank); this is compatible with the prevalence of long-term relations with customers (70% versus 30% occasional);

– The fact that competitors have not changed their prices, which suggests the presence of a co-ordination failure (4th rank).

Why do prices change infrequently? (2)• Both reasons suggest imperfect competition is a crucial factor in explaining price stickiness.

• Greater competition implies larger frequency of price changes:

– Consumer prices: Type of outlet matters; higher frequency in supermarkets, lower in corner shops

– Producer prices: Import penetration matters; more open sectors change prices more frequently.

– Survey evidence: firms in more competitive sectors change prices more frequently

Why do prices change infrequently? (2)Frequency of price reviews by degree of perceived competitiona

0

10

20

30

40

50

60

70

80

Very low competition Low competition High competition Very high competition

More than twelve Between four and twelve Less than four

Frequency of price changes by degree of perceived competitiona

05

101520253035404550

Very low competition Low competition High competition Very high competition

More than four Two or three One Incidental or none

Why do prices change infrequently? (2)• Other factors:

– Attractive pricing matters, especially in the retail sector, but also at the producer level.

– Seasonal effects:

• e.g. a large share of price changes takes place in January, particularly in services.

– Higher and more variable inflation increases the frequency of price adjustment.

Other aspects of price setting• Some survey evidence of rule-of-thumb

behaviour

• Robust survey evidence on asymmetries in the reasons for price changes:

– Cost-push factors are more important in driving prices up

– Changes in demand are more important in driving prices down

Belgium Spain Luxembourg PortugalShare of firms applying

a rule of thumb 37 33 33 25

Other aspects of price setting

• Not much evidence of synchronisation

– but synchronisation measures go up as one analyses more similar products or geographically more closer outlets.






Overview

Policy implications (1)

• Inflation persistence is moderate in stable, low inflation regimes and has fallen over the most recent period:

– Makes it easier to stabilise inflation following shocks;

– Reflects the benefits from well-anchored inflation expectations;

– Monitoring and anchoring inflation expectations is crucial in retaining those benefits


0 5 10 15 20-0.8

-0.6

-0.4

-0.2

0

0.2Output gap

LowMediumHigh

0 5 10 15 20-0.1

0

0.1

0.2

0.3

0.4Interest rate

0 5 10 15 20-0.5

0

0.5

1

1.5Inflation

0 5 10 15 20-1

-0.5

0

0.5Real rate

Response to a cost-push shockunder different degrees of intrinsicInflation persistence (Smets-Wouters)


• Relatively high degree of price stickiness implies:

– that inflation is likely to be less responsive to changes in real marginal costs and output gaps;

– that deviations of inflation from the objective are more costly to control;

• Supports the case for monetary policy to focus on maintaining price stability;


0 5 10 15 20-0.3

-0.2

-0.1

0

0.1Output gap

LowMediumHigh

0 5 10 15 20-0.1

0

0.1

0.2

0.3

0.4Interest rate

0 5 10 15 20-0.5

0

0.5

1Inflation

0 5 10 15 20-0.4

-0.2

0

0.2

0.4

0.6Real rate

Response to a cost-push shockunder different degrees of pricestickiness (Smets-Wouters)


• Heterogeneity in the degree of price stickiness

– Important to monitor the nature and the source of inflation shocks, as it helps improving inflation projections;

– Literature suggests to target a price index that assigns a relatively larger weight to the more sticky sectors in order to avoid excessive adjustment of these sectors.

– This suggests a more prominent role for indices such as the HICP excluding energy and food prices;

– But there are important implementation issues: measurement, transparency and communication, accommodation of structural inefficiencies?


• Large share of price changes are price decreases

– Reduces the need for an inflation buffer to facilitate relative price adjustments, but…

– … price decreases are relatively rare in services, which have a large share in the HICP and the economy as a whole …

– … one possible reason could be downward nominal wage rigidities!


• Prices are more sticky than in the United States:

– A full analysis of the underlying reasons is still outstanding, …

– … but differences in the degree of competition in product and labour markets are a likely explanation

• In particular, the services sectors are characterised by a very high degree of nominal price rigidities and downward price stickiness

– This may be related to more important role of administrative prices; less competition in many services sectors; and the greater importance of sluggish wage adjustment and downward wage rigidity.


• Structural reforms in services and labour markets that are geared at accelerating price and wage adjustments would:

– Facilitate the stabilisation of inflation following cost-push shocks;

– Accelerate necessary relative price adjustments across sectors and countries following asymmetric shocks, lowering the persistence of inflation differentials.

Issues for further research

• There is a lot of food for thought in the findings of the IPN:

– The findings of state dependence in price setting and the important role of strategic interactions amongst price setters and their customers deserve further investigation

• Two important priorities:

– What drives the differences in price setting between the euro area and the United States

– The role of wages and wage rigidities …

• A new Eurosystem Research Network (Wage Dynamics Network) is up and running …

Documents

National Bank of Belgium 12-13 October, 2006