“Inflation Targeting: The Experience of Emerging Markets” N Batini (RES, WEO), D Laxton (RES, EM) With support from M Goretti (RES, WEO). Research Assistance:

““Inflation Inflation Targeting: Targeting:

The Experience of The Experience of Emerging Markets”Emerging Markets”

N Batini N Batini (RES, WEO), (RES, WEO), D Laxton D Laxton (RES, (RES, EM)EM)

With support from With support from M Goretti M Goretti (RES, (RES, WEO). Research Assistance:WEO). Research Assistance: N N

CarcenacCarcenac

FACTSFACTS

T very popular monetary policy T very popular monetary policy strategystrategy

21 countries (of which 8 advanced 21 countries (of which 8 advanced and 13 emerging markets) are now and 13 emerging markets) are now TersTers

Many more are thinking to adopt Many more are thinking to adopt TT

LITERATURELITERATURE

Recently a few papers have looked Recently a few papers have looked at whether at whether T improves macro-T improves macro-performance (“performance (“T matters”) in the T matters”) in the context of context of industrial economiesindustrial economies

YesYes: : Kuttner and Posen (2001), Levin et Kuttner and Posen (2001), Levin et al (2004), Hyvonen (2004), Truman (2004)al (2004), Hyvonen (2004), Truman (2004)

NoNo: : Ball and Sheridan (2003)Ball and Sheridan (2003)

MOTIVATIONMOTIVATION

Is it a good idea, from a macro Is it a good idea, from a macro perspective, to adopt perspective, to adopt T?T?

Are there any other benefits or costs Are there any other benefits or costs to to T?T?

Are there preconditions to adopt Are there preconditions to adopt T?T?

What should the Fund advice on What should the Fund advice on T?T?

MOTIVATIONMOTIVATION

Is it a good idea, from a macro Is it a good idea, from a macro perspective, to adopt perspective, to adopt T?T?

Are there any other benefits/costs to Are there any other benefits/costs to T?T?

Are there preconditions to adopt Are there preconditions to adopt T?T?

What should the Fund advice on What should the Fund advice on T?T?

METHODOLOGYMETHODOLOGY

Use econometric tools to answer Use econometric tools to answer questions based both on questions based both on surveysurvey and and “hard” data“hard” data

Look at Look at emerging market emerging market economieseconomies

METHODOLOGY (CONT.)METHODOLOGY (CONT.)

Survey contains over 130 questionsSurvey contains over 130 questions

3 parts: 3 parts: institutional, economic institutional, economic andand political economypolitical economy facts facts

Asked in person to all emerging Asked in person to all emerging market market TersTers

Email and phone for other Email and phone for other Ters and Ters and nonnon

WHAT IS WHAT IS T?T?

T is an operational framework for T is an operational framework for monetary policy aimed at attaining monetary policy aimed at attaining price stabilityprice stability

Contrary to alternative strategies, Contrary to alternative strategies, notably money or exchange rate notably money or exchange rate targeting, targeting, T involves targeting T involves targeting inflation inflation directlydirectly

WHAT IS IT?WHAT IS IT?

2 main characteristics:2 main characteristics:

1.1. Unique targetUnique target, specifying , specifying numericallynumerically the objective of price the objective of price stability in the form of a stability in the form of a level or a level or a range for annual inflationrange for annual inflation

2.2. The The inflation forecastinflation forecast is the de is the de facto target variablefacto target variable

OTHER (ANCILLARY) OTHER (ANCILLARY) T T CHARACTERISTICSCHARACTERISTICS

Transparency (goal vs. operational)Transparency (goal vs. operational) CommunicationCommunication AccountabilityAccountability

T VERSUS MONETARY T VERSUS MONETARY POLICY IN THE US, JAP AND POLICY IN THE US, JAP AND

THE EA ? THE EA ?

USUS, , JAPJAP: : nono numerical target on numerical target on inflationinflation

EAEA: Inflation numerical objective, : Inflation numerical objective, but alsobut also reference value for M3 reference value for M3 growth. Not growth. Not as great an emphasisas great an emphasis on on inflation projection as ITers (“two” inflation projection as ITers (“two” pillars: economic pillars: economic andand monetary monetary analysis)analysis)

Proponents say: with Proponents say: with T, T,

Unique clear objective and transparencyUnique clear objective and transparency speed learningspeed learning & & help anchor help anchor expectationsexpectations faster & more durably faster & more durably

Thanks to medium-term orientation, Thanks to medium-term orientation, T T grants grants more flexibilitymore flexibility (milder on (milder on output gap variability). This requires output gap variability). This requires greater accountability (“constrained greater accountability (“constrained discretion”)discretion”)

Lower cost of policy failureLower cost of policy failure

T better than PEGS…T better than PEGS…

Milder on business cycleMilder on business cycle (exchange rate (exchange rate targeting is price level targeting on one targeting is price level targeting on one individual price)individual price)

Target is controllableTarget is controllable under under T, not under T, not under pegs (domestic versus international pegs (domestic versus international reputational equilibrium)reputational equilibrium)

T (as other flex regimes) T (as other flex regimes) minimizes minimizes negative consequences of exchange rate negative consequences of exchange rate volatilityvolatility on real activity on real activity

T better than MONEY T better than MONEY TARGETS…TARGETS…

Better at anchoring expectations Better at anchoring expectations (single target, mandate more clear (single target, mandate more clear and monitorable)and monitorable)

More flexible (longer horizon)More flexible (longer horizon) Optimal money growth time-varying. Optimal money growth time-varying.

Optimal inflation target static.Optimal inflation target static.

Critics say: withCritics say: with T,T,

Too little discretion, growth Too little discretion, growth unnecessarily restrainedunnecessarily restrained

Too much discretion—cannot help Too much discretion—cannot help build credibilitybuild credibility

Implies exchange rate neglectImplies exchange rate neglect It cannot work were ‘preconditions’ It cannot work were ‘preconditions’

are poorare poor

Regional Average Annual Inflation Rate (percent)Regional Average Annual Inflation Rate (percent)

So is So is T BETTER or T BETTER or WORSE? WORSE?

Average Annual Inflation Rate

0

5

10

15

20

25

30

35

40

19901 19921 19941 19961 19981 20001 20021 20041

IT

Non ITers

Volatility Annual Inflation Rate

0

5

10

15

20

25

30

35

40

19901 19921 19941 19961 19981 20001 20021 20041

IT

Non ITers

Inflation and growth performanceInflation and growth performance

Ters Non- Ters

Non-inflation Targeters: Annual Inflation Rate(pre inflation <40)

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20 25 30

Average

Vola

tilit

y

Pre IT Post IT

Non-inflation Targeters: Annual Growth Rate (pre inflation<40)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

-2 0 2 4 6 8

Average

Vola

tilit

y

Inflation Targeters: Annual Inflation Rate(pre inflation <40)

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20 25 30 35 40

Average

Vola

tilit

y

Pre IT Post IT

Inflation Targeters: Annual Growth Rate(pre inflation <40)

0

1

1

2

2

3

3

4

-2 0 2 4 6 8 10

Average

Vola

tilit

y

How does How does TT affect affect

macroeconomimacroeconomic outcomes?c outcomes?

Very hard to answer for Very hard to answer for industrial economies…industrial economies…

Small sample. Small sample. 7 adopters in early-mid 90s, 2 of which 7 adopters in early-mid 90s, 2 of which

joined the Euro area; 3 more in ‘99-’01.joined the Euro area; 3 more in ‘99-’01. Limited set of “control” countries.Limited set of “control” countries.

Many candidates joined the Eurozone.Many candidates joined the Eurozone. Not much room for improvement. Not much room for improvement. Most Most nonnon--T ers did better in the T ers did better in the

1990s. 1990s.

What can EM countries What can EM countries tell us?tell us?

Larger sample:Larger sample: 13 emerging-market adopters since 13 emerging-market adopters since

19971997 10 of these prior to 200210 of these prior to 2002

Larger set of potential “control” Larger set of potential “control” countries.countries.

Much more room for improvement in Much more room for improvement in most cases.most cases.

Assessing the EM Assessing the EM experience is also difficult…experience is also difficult… ShortShort post- post-T sampleT sample

Most adopted between 1999 and 2001Most adopted between 1999 and 2001 Extremely heterogeneous sampleExtremely heterogeneous sample

Lots of things were going on Lots of things were going on besidesbesides T T Most non-Most non-Ter EM countries have Ter EM countries have

alsoalso done better in recent years. done better in recent years.

Bottom line in advanceBottom line in advance

Emerging-market Emerging-market Ters Ters did did do do better than comparable non-better than comparable non-Ters.Ters. Lower inflationLower inflation More stable inflationMore stable inflation More anchored long-run inflation More anchored long-run inflation

expectationsexpectations Lower output volatilityLower output volatility

T beats (successful) pegs.T beats (successful) pegs.

The empirical methodThe empirical method

Step 1: partition the sample into Step 1: partition the sample into “pre” and “post” periods.“pre” and “post” periods.

Step 2: select the sample of Step 2: select the sample of countries.countries.

Step 3: compare average “pre” to Step 3: compare average “pre” to average “post” performance.average “post” performance.

How to partition the How to partition the sample?sample?

Scheme “pre” “post”Baseline 1971* to –1 to 2004 T

1971* to ’99 2000 to ’04 non-T

Time 1994 to ‘96 2002 to ’04 allperiods

Actual 1971* to –1 to 2004 T

dates 1971* to s–1 s to 2004 non-T

* Or beginning of data, if after this date = T adoption date s = non-Ters’ most recent regime change

How to select the How to select the sample?sample?

42 countries:42 countries: 13 emerging market 13 emerging market TersTers Comparable non- Comparable non- T EM countriesT EM countries

22 emerging market countries (in 22 emerging market countries (in JPMorgan EMBI index)JPMorgan EMBI index)

7 additional countries:7 additional countries: Botswana, Costa Rica, Ghana, Guatemala, Botswana, Costa Rica, Ghana, Guatemala,

India, Jordan, TanzaniaIndia, Jordan, Tanzania

Basic empirical Basic empirical specificationspecification

Xi,t = [ T di,t + N (1– di,t) ] + (1– ) Xi,t–1

XX = performance metric: = performance metric: , SD(, SD(), SD(), SD(yy)) dd = = T dummyT dummy Ters “revert” to Ters “revert” to TT , non-, non-Ters to Ters to NN

= “speed of reversion”= “speed of reversion”

Letting 0 = N, 1= (T - N) and b = - ,

Xi,t = 0 + 1 di+ bXi,t–1 +ei

The Ball-Sheridan The Ball-Sheridan regressionregression

Xi,t = [ T di,t + N (1– di,t) ] + (1– ) Xi,t–1

Xi,2 – Xi,1 = T di,t + N (1– di,t) – Xi,1

Xi,2 – Xi,1 = a0 + a1 di,t + b Xi,1 + ei

= –b T = (a0 + a1 )/N= a0/

H0: a1 = 0 level of X is unaffected by T

Variables T dummy variable

–4.820

SD() –3.638

SD( y-y*) –0.010

SD (growth) –0.633

Baseline resultsBaseline results

Significant at 10% level, 5% level, 1% level

Estimates of coefficient on IT dummy


5-year forecast, level –2.672

6-10-year forecast, level –2.076

5-year forecast, SD –2.185

6-10-year forecast, SD –1.737

Inflation expectationsInflation expectations



EMP index – 0.340

Reserves volatility -16.333

Exchange rate volatility –11.090

Interest rate volatility –5.025

Crises proclivityCrises proclivity**


* Similar tests on other countries - with flexible exchange rates but different monetary regimes - show either a not significant effect or an even higher crisis likelihood.

1. Sample partitioning

2. High-inflation countries (>40 %)

3. Low-income countries (WB)

4. Countries not incl. in EMBI index

5. Severely indebted countries (WB)

6. Fixed exchange rate regimes

7. Different degrees of fiscal discipline

Robustness ChecksRobustness Checks

1. Sample partitioning

2. High-inflation countries (>40 %)

3. Low-income countries (WB)

4. Countries not incl. in EMBI index

5. Severely indebted countries (WB)

6. Fixed exchange regimes

7. Different degrees of fiscal discipline

Robustness ChecksRobustness Checks

Coefficient on dummy for:

Variables T ERT

–4.820 – 0.084

SD() –3.638 1.124

SD(y-y*) –0.010 0.030

Comparing Alternative Comparing Alternative Regimes:Regimes:

Exchange Rate Targets*Exchange Rate Targets*


* We include in this category conventional pegs, currency boards and countries with another currency as legal tender

Conclusion on macro Conclusion on macro performanceperformance

IT has improved macro outcomes in IT has improved macro outcomes in emerging market economies:emerging market economies:

IT confers significantly larger IT confers significantly larger benefits of an exchange rate peg, benefits of an exchange rate peg, and without the fragilityand without the fragility

The role of The role of institutional institutional

and structural and structural conditionsconditions

Institutional and structural Institutional and structural factorsfactors

To what extent does To what extent does T require T require specific institutional and/or specific institutional and/or structural conditions to be met?structural conditions to be met? Conventional wisdom: Conventional wisdom: T requires T requires

rigorous preconditions!rigorous preconditions! Does the adoption of Does the adoption of T catalyze T catalyze

favorable institutional and/or favorable institutional and/or structural change?structural change?

What are these factors?What are these factors?

Institutional independenceInstitutional independence

Technical infrastructureTechnical infrastructure

Financial system healthFinancial system health

Economic structureEconomic structure

1. Institutional 1. Institutional independenceindependence

Operational independenceOperational independence Control over rate settingControl over rate setting

Central bank autonomyCentral bank autonomy No obligation to finance government No obligation to finance government

expendituresexpenditures Fiscal discipline (low gov. balance & debt)Fiscal discipline (low gov. balance & debt) No (threat of) interference from No (threat of) interference from

governmentgovernment A clear, focused mandateA clear, focused mandate

2. Technical 2. Technical infrastructureinfrastructure

Forecasting capabilityForecasting capability Inflation forecast is central to Inflation forecast is central to T T

Analytical & modeling capabilityAnalytical & modeling capability Needed to assess likely impact of policy Needed to assess likely impact of policy

actionsactions

Data availability & qualityData availability & quality

3. Financial system 3. Financial system healthhealth

Sound banking sectorSound banking sector

Reasonably well-developed financial Reasonably well-developed financial marketsmarkets

Limited degree of currency Limited degree of currency mismatchmismatch Minimizes likely conflict with monetary Minimizes likely conflict with monetary

policy objectivespolicy objectives

4. Economic structure4. Economic structure

Not too sensitive to exchange rate & Not too sensitive to exchange rate & commodity price shockscommodity price shocks

Little or no dollarizationLittle or no dollarization

Little trade openness (less exposed Little trade openness (less exposed to external shocks and spillovers)to external shocks and spillovers)

How to measure How to measure institutional and structural institutional and structural

characteristics?characteristics? Data from our survey of Data from our survey of Ters and Ters and

nonnon- Ters.Ters. A wealth of detail and anecdotes—but a A wealth of detail and anecdotes—but a

challenge to “quantify”.challenge to “quantify”. Caveat: reliability of self-reported data!Caveat: reliability of self-reported data!

Supplemented with more Supplemented with more conventional “hard” data.conventional “hard” data.

Initial conditions prior to Initial conditions prior to adopting adopting TT

PhilippinesIsrael

Czech RepublicPeru

HungaryKoreaBrazilChile

ThailandPoland

ColombiaSouth Africa

Mexico

0 1 2 3 4

New Zealand

Iceland

Australia

Norway

Canada

Sweden

United Kingdom

Switzerland

0 1 2 3 4

Financial system healthTechnical infrastructure

Figure 4.3. Initial Conditions Prior to Adopting Inflation Targeting(0 = poor; 1 = ideal; for each of the four categories of initial conditions)

Source: IMF staff calculations.

Emerging Markets

Economic structureInstitutional independence

Industrial Countries

Idea

l con

ditio

nsId

eal c

ondi

tions

Most of the inflation targeters had poor initial conditions prior to the adoption of inflation targeting.

PhilippinesIsrael

Czech RepublicPeru

HungaryKoreaBrazilChile

ThailandPoland

ColombiaSouth Africa

Mexico

0 1 2 3 4

New Zealand

Iceland

Australia

Norway

Canada

Sweden

United Kingdom

Switzerland

0 1 2 3 4

Financial system healthTechnical infrastructure

Figure 4.3. Initial Conditions Prior to Adopting Inflation Targeting(0 = poor; 1 = ideal; for each of the four categories of initial conditions)

Source: IMF staff calculations.

Emerging Markets

Economic structureInstitutional independence

Industrial Countries

Idea

l con

ditio

nsId

eal c

ondi

tions

Most of the inflation targeters had poor initial conditions prior to the adoption of inflation targeting.

Do preconditions (or lack Do preconditions (or lack thereof) affect thereof) affect Ters’ Ters’

performance?performance? No.No. We constructed preconditions We constructed preconditions

proxies, based on survey + “hard” proxies, based on survey + “hard” data.data.

These turn out to be insignificant in These turn out to be insignificant in Ball-Sheridan-style regressions for Ball-Sheridan-style regressions for Ters.Ters.

Post-adoption progress on Post-adoption progress on conditions however maybe conditions however maybe

vital...vital...

Pre-adoption:Pre-adoption:

Emerging Markets

0 1 2 3 4

Brazil

Chile

Colombia

Mexico

Peru

Thailand

Korea

Philippines

Poland

Czech Republic

Hungary

South Africa

Israel

Technical Infrastructure Financial System HealthInstitutional independence Economic structure

Post-Adoption:Post-Adoption:

Emerging Markets

0 1 2 3 4

Brazil

Chile

Colombia

Mexico

Peru

Thailand

Korea

Philippines

Poland

Czech Republic

Hungary

South Africa

Israel

Technical Infrastructure Financial System HealthInstitutional independence Economic structure

ConclusionsConclusions

T “matters” for EM economies.T “matters” for EM economies.

Preconditions should not be a Preconditions should not be a serious obstacle to adopting serious obstacle to adopting T T

Prospective Prospective Ters look a lot alike Ters look a lot alike current current Ters at time of adoptionTers at time of adoption

Documents

“Inflation Targeting: The Experience of Emerging Markets” N Batini (RES, WEO), D Laxton (RES, EM) With support from M Goretti (RES, WEO). Research Assistance: