Targeting inflation in Asia and the Pacific Andrew Filardo* Hans Genberg** *Bank for International Settlements **Hong Kong Monetary Authority The views.

Targeting inflation in Asia and the Pacific

Andrew Filardo*Hans Genberg**

*Bank for International Settlements**Hong Kong Monetary Authority

The views expressed are those of the authors and do not necessarily represent those of the Bank for International

Settlements or the Hong Kong Monetary Authority.

2

Inflation targeting vs.targeting inflation

• Inflation performance in the region has been good

• True for inflation targeting central banks but also for others

• Targeting inflation is important

• Inflation objective can be achieved in more ways than one.

3

Agenda

• A review of policy objectives and strategies of central banks in Asia and the Pacific

• Evolution of central bank independence and other aspects of governance

• The inflation record in IT and non-IT economies– Inflation rates– Inflation persistence– Properties of inflation forecasts– Effects of the recent commodity price cycle

• Policy implications

4

Central bank policy objectivesJurisdiction Policy objective

Australia Price stability

China Value of the currency

Hong Kong SAR Exchange rate stability

India Price stability and adequate credit supply

Indonesia Price stability and exchange rate stability

Japan Price stability

South Korea Price stability

Malaysia Price stability and exchange rate stability

New Zealand Price stability

Philippines Price stability

Singapore Price stability

Thailand Price stability

5

Inflation targeting central banksStarting year Targeting arrangement

Australia 1993 Target for headline CPI consumer price inflation of 2-3 per cent per annum on average over the business cycle

Indonesia 2000 Tnflation target for 2008, 2009, and 2010 is 5±1%, 4.5±1%, and 4±1% for y-o-y CPI inflation

Korea 1999 Target range of 3±0.5% in terms of 3-year average of annual CPI inflation

New Zealand 1990 Target range of 1 to 3% on average, over the medium term, defined in terms of the All Groups Consumers Price Index (CPI)

Philippines 2002 Target range of 3.5±1% (2009), 4.5±1% (2010) for the avg year-on-year change in the CPI over the calendar year.

Thailand 2000 Target range of 0-3.5% for quarterly average of core inflation.

6

Non-IT central banks

Targeting arrangement

China Reference to money growth targets

Hong Kong Currency board: target range centered on HKD 7.8 = USD 1

India Multiple objectives: price stability understanding - containing the perception of inflation in the range of 4.0-4.5% so that an inflation rate of 3.0% becomes the medium term objective.

Japan Medium- to long-term price stability expressed in terms of year on year rate of change in the CPI (approximately between 0 and 2%).

Malaysia Multiple indicators and instruments

Singapore As of mid-2009, zero percent appreciation of the undisclosed S$NEER policy band

7

Improved governance structures

• Greater legal and political independence

• Greater ability to carry out independent exchange rate policy

• Increased ability to carry out monetary policy independently of fiscal policy

• Increased transparency and accountability

• For IT central banks: Greater emphasis on price stability as the primary objective

8

Improved governance, 1996-2005

9

Average inflation rates

0.001.00

2.003.004.00

5.006.007.00

8.009.00

Old' IT New-IT Non-IT

1996-00

2001-08

The inflation record

10

Some more detail

• Inflation persistence

• Properties of inflation forecasts

• Impact of commodity price cycle

11

Inflation persistence

• Inflation is less persistent in IT economies– Smaller AR coefficient– Transitory shocks relatively more important

12

Changing AR(1) inflation persistence

1t t tC

.

13

Changing IMA(1) inflation persistence

The greater | | , the greater the proportion of the inflation variance accounted for by the transitory component.

14

Properties of inflation forecasts

• Use Consensus Forecasts to investigate whether:– Better central bank governance reduces the

dispersion of inflation forecasts– Inflation forecasts are more homogeneous in

IT economies?

15

The Kullback-Leibler divergence measure

dxxu

xpxpD LK

)(

)(log)(

0%

20%

40%

60%

80%

100%

-2 0 2 4 6

Inflation

Pro

ba

bilit

y

p(x) q(x) u(x)

16

i i i i it t j CBGI t tKLn C KLc CBGI e

Table 4The impact of the changes in CBGI

on the cross-sectional distribution of inflation forecasts in Asia-Pacific

Index β θCBIG Nobs R2

Overall0.62(5.3)

0.30(1.2)

116 0.68

Monetary policy

independence

0.62(5.3)

0.30(1.2)

116 0.67

Notes: 1) Using January sample for 1991 to 2005; 2) estimated using fixed effects and t-statistics in parentheses based on robust standard errors.

17

11 11

1 1

i i i i n i i n i it t j IT t AIT t IT t t j AIT t t j t

n n

KLn C KLc I I I KLc I KLc e

The impact of the adoption of inflation targeting on the cross-sectional distribution of inflation forecasts in Asia-Pacific

Using January sample

(1) (2) (3) (4) (5) (6) (7) (8)

KL in levels ΔKL

β .62 (6.8) .61 (6.8) .59 (6.6) .57 (6.3) .60 (5.2) .60 (5.1) .54 (3.4) .53 (3.4)

γIT .21 (4.2) -.31 (-2.4) -.18 (-0.9) .09 (.14) .06 (.52)

γAIT .05 (3.4) .10 (4.4) .12 (7.8) .06 (.49) .02 (3.3)

βIT -.12 (-1.1) -.26 (-1.0) .37 (1.8) .39 (1.8)

βAIT .03 (.52) -.12 (-4.1)

-.12 (-4.2)

Nobs 163 163 163 163 163 163 149 149

0.66 0.66 0.66 0.66 0.66 .67 .32 .35

18

Impact of commodity price cycle

Asia Mature economies

Headline

Excluding food and energy

19

20

Summary or empirical evidence

• Admirable inflation performance

• Difficult to document differences between explicit IT economies and non-IT economies

Why?

• Common shocks → Common outcomes

• Social consensus about importance of inflation control.– Central bank objectives– Governance structures

21

Policy challenges 1

• How to reconcile primary focus on inflation with increasing concern in society about other objectives?– Volatility of capital flows/sudden stops– Exchange rate misalignments– Financial stability and economic growth

• Policy frameworks must be such as to prevent these factors from undermining the consensus in favour of inflation control.

22

Policy challenges 2

• Need as many independent instruments as you have objectives.– Impeccable logic in theory but:

• In some cases the trade-off between objectives may translate into a trade-off between inflation in the near term vs. inflation later

• What if other instruments are not used?• There may still be a case for coordination.

• Lexicographic approach– Assumes independence between objectives– Not what central banks have been observed to do

recently

23

Policy challenges 3

• Practice– Expanded perspectives. Pay attention to

emerging imbalances• Two pillars (ECB), two perspectives (BoJ)

• Theory:– State-dependent preferences?, escape

clauses?

• Work in progress

24

Forecast accuracy

• Forecast Accuracy = F(governance indicator, IT dummy, controls)

• Preliminary results:– Better governance → smaller forecast errors– IT → smaller forecast errors– Governance indicator drives out IT dummy