Structural Reforms in Brazil

1
Inflation Targeting in Brazil Lessons and
Challenges
November 2002
2
The paper assesses the first three years of
inflation targeting (IT) in Brazil (1999-2002)

• Overview of the three years
• Challenges in the inflation targeting
• - construction of credibility
• - change in relative prices
• - exchange rate volatility
• Methodology to estimate effects of shocks
• Institutional design of IT core, escape clauses,
  tolerance intervals, targets

3
Main Results of Estimations

• Inflation targets -gt important coordinator of
  expectations
• CB -gt reacted strongly to inflation expectations
• Degree of persistence of inflation -gt reduction
• Volatility inflation and output -gt reduction

4
Main Results (cont.)

• Pass-through of Exchange Rate
• - to IPCA between 10 and 20
• - to administered prices 2 times higher than to
  market prices
• - explains 38-48 of inflation (2001-Aug.2002)

5
Overview of the three years
6
Macroeconomic Policy

• Inflation Targeting
• Floating Exchange Rate
• Change in the fiscal regime
• primary surplus (GDP)
• 0.0 1998 3.7 2001
• 3.2 1999 3.5 Aug 2002 (12-m)
• 3.5 2000

7
IT regime successful, highly important for
macroeconomic stabilization

• 1999 and 2000 targets met
• 2001 and 2002 targets not met - several shocks
  hitting the economy

12
10
8.94
7.67
8
5.97
6
4
2
0
1999
2000
2001
2002
2003
2004
8
Volatilities

• IT period lower volatility of inflation, GDP,
  and interest rate

Interest Rate
Inflation Rate
Output
Period
Coefficient
Coefficient
Average
Standard
Coefficient
Average
Standard
Average
Standard
of
of
(per year)
Deviation
of Variation
(per year)
Deviation
(per year)
Deviation
Variation
Variation
Real Plan Before
Inflation Targeting
199404 - 199902
10.3
9.2
0.9
2.0
6.3
3.2
35.4
14.1
0.40
199601 - 199902
5.8
4.8
0.8
2.0
5.2
2.5
28.2
6.0
0.21
Inflation Targeting
199903 - 200202
7.1
3.0
0.4
2.4
3.5
1.5
18.0
1.4
0.08
9
Construction of Credibility
10
Reaction function of Central Bank

• Interest rate reacts strongly to expected
  inflation
• Monetary policy is conducted in a forward-looking
  basis and is tough against inflation

11
Estimation of Reaction Function of Central Bank
Dependent Variable Selic Interest Rate (Monthly
Average)
Coefficients and standard errors
Regressors
With Inflation Report
With Market Inflation
Inflation Expectations
Expectations
Constant
17.50
17.56
16.48
16.66
(0.36)
(0.49)
(0.64)
(0.70)
Interest Rate (t-1)
0.76
1.04
0.72
1.36
(0.07)
(0.13)
(0.08)
(0.18)
Interest Rate (t-2)
-0.20
-0.56
(0.08)
(0.15)
Deviation of Expected Inflation Rate from Target
1.77
1.85
1.75
1.42
(0.85)
(1.21)
(0.67)
(0.72)
Output Gap (t-1)
-0.36
-0.38
-0.19
-0.10
(0.09)
(0.13)
(0.11)
(0.13)
R-squared
0.9287
0.9415
0.9270
0.9538
Adjusted R-squared
0.9220
0.9340
0.9186
0.9464
LM Test for Autocorrelation of Residuals
(p-values)
1 lag
0.0352
0.4877
0.0056
0.7313
4 lags
0.2149
0.6671
0.0738
0.5637
Notes Standard error in parantheses. , , and
indicate the coefficient is significant at
10, 5, and 1 level, respectively.
12
Estimation of Reaction Function of Central Bank
Dependent Variable Gap of Selic Interest Rate
(Monthly Average)
Coefficients and standard errors
Regressors
With Inflation Report
With Market Inflation
Inflation Expectations
Expectations
Constant
-1.51
-1.27
-3.29
-3.53
(0.36)
(0.36)
(0.54)
(0.65)
Gap of Interest Rate (t-1)
0.81
1.08
0.71
1.34
(0.06)
(0.09)
(0.08)
(0.19)
Gap of Interest Rate (t-2)
-0.25
-0.54
(0.06)
(0.15)
Deviation of Expected Inflation Rate from Target
4.98
4.24
3.71
3.63
(0.91)
(0.77)
(0.58)
(0.68)
Output Gap (t-1)
-0.30
-0.34
-0.03
0.07
(0.12)
(0.11)
(0.10)
(0.13)
R-squared
0.9654
0.9768
0.9694
0.9798
Adjusted R-squared
0.9622
0.9738
0.9658
0.9765
LM Test for Autocorrelation of Residuals
(p-values)
1 lag
0.1309
0.4205
0.0079
0.4261
4 lags
0.0807
0.4993
0.0456
0.4331
Notes Standard error in parantheses. , , and
indicate the coefficient is significant at
10, 5, and 1 level, respectively.
13
Inflation Expectations under control
12-Month Ahead Expected Inflation and Inflation
Target
9
8
7
6
5
p.a.
4
3
2
1
0
Jul-00
Jul-01
Jul-02
Jan-00
Jan-01
Jan-02
Mar-00
Sep-00
Mar-01
Sep-01
Mar-02
Sep-02
May-00
Nov-00
May-01
Nov-01
May-02
Central Target
Upper Limit
Lower Limit
Inflation Expectation
14
Reaction Function of Inflation Expectations

• Inflation expectations respond to
• - interest rate (positively)
• - inflation rate target (positively)
• - past inflation is not significant.
• Inflation expectations are forward looking and
  the inflation targets have an important role

15
Estimation of Reaction Function of Inflation
Expectations
Dependent Variable Market Inflation Rate
Expectations (Adjusted)
Coefficients and standard errors
Regressors
I
II
III
IV
Constant
-4.23
-4.75
-4.22
-4.92
(1.30)
(1.31)
(1.32)
(1.35)
Market Inflation Rate Expectations (t-1)
0.24
0.48
0.27
0.50
(0.18)
(0.19)
(0.18)
(0.19)
Market Inflation Rate Expectations (t-2)
-0.39
-0.40
(0.15)
(0.15)
Interest Rate (t-1)
0.27
0.29
0.25
0.27
(0.07)
(0.07)
(0.08)
(0.08)
Inflation Rate Target (12-Month Ahead)
0.74
0.96
0.70
0.95
(0.22)
(0.24)
(0.23)
(0.24)
12-Month Inflation Rate (t-1)
0.06
0.08
(0.10)
(0.11)
R-squared
0.8978
0.9030
0.8993
0.9053
Adjusted R-squared
0.8855
0.8861
0.8826
0.8838
LM Test for Autocorrelation of Residuals
(p-values)
1 lag
0.0663
0.5196
0.0652
0.7730
4 lags
0.0160
0.1333
0.0149
0.1316
16
Change in the inflation dynamics

• Simple Aggreggate Supply Curve
• - IT period decrease in the degree of
  persistence in inflation

17
Estimation of Aggregate Supply Curve
Dependent Variable Monthly Inflation Rate
Coefficients and standard errors
Regressors
First
Second
Third
Specification
Specification
Specification
Constant
0.74
0.81
0.79
(0.36)
(0.35)
(0.32)
DummyConstant
0.38
0.56
0.39
(0.14)
(0.15)
(0.14)
Inflation Rate (t-1)
0.74
0.81
0.61
(0.09)
(0.12)
(0.13)
DummyInflation Rate(t-1)
-0.59
-0.58
-0.41
(0.20)
(0.21)
(0.19)
Inflation Rate (t-2)
-0.12
-0.09
(0.12)
(0.12)
DummyInflation Rate (t-2)
-0.30
-0.24
(0.20)
(0.19)
Unemployment (t-1)
-0.09
-0.10
-0.10
(0.05)
(0.05)
(0.05)
Exchange Rate Change (t-1)
0.10
(Twelve-Month Average)
(0.04)
Dummy for 200007
1.08
(0.33)
R-squared
0.6431
0.6766
0.5537
Adjusted R-squared
0.6269
0.6538
0.5055
LM Test for Autocorrelation of Residuals
(p-values)
1 lag
0.1857
0.8353
0.5454
4 lags
0.0040
0.1693
0.1081
18
Change in relative prices
19
Relative increase in the administered or
monitored prices

• The group includes gasoline, cooking gas,
  electricity, telephone, urban bus
• 30 of the consumer price index
• Dynamics are different from the other prices
• - international prices
• - greater pass-through from exchange rate
• - stronger backward-looking behavior

20
Ratio of Administered Prices to Market Prices
1.6
1.4
1.2
1.0
0.8
Jul-92
Jul-93
Jul-94
Jul-95
Jul-96
Jul-97
Jul-98
Jul-99
Jul-00
Jul-01
Jan-92
Jan-93
Jan-94
Jan-95
Jan-96
Jan-97
Jan-98
Jan-99
Jan-00
Jan-01
Jan-02
21
Level of Prices IPCA and selected items
Dec-981
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
Apr-99
Jun-99
Oct-99
Apr-00
Jun-00
Oct-00
Apr-01
Jun-01
Oct-01
Apr-02
Jun-02
Dec-98
Feb-99
Dec-99
Feb-00
Dec-00
Feb-01
Dec-01
Feb-02
Aug-99
Aug-00
Aug-01
IPCA
Cooking Gas
Electricity
Gasoline
Fixed Telephone
Urban Bus
22
Granger Causality Test Relative Prices and IPCA
Sample 199412 200206
Null Hypothesis
1 lag
3 lags
F-statistic
P-value
F-statistic
P-value
Relative Prices do not Granger Cause Inflation
Rate
2.89
0.0926
3.15
0.0293
Inflation Rate does not Granger Cause Relative
Prices
16.71
0.0001
5.35
0.0020
23
Exchange Rate Volatility
199907 200206 Average Monthly Increase 1.2
p.m. s.e. 3.6
24
Exchange Rate Level (R/US)
3.50
3.00
2.50
2.00
1.50
1.00
Jul-98
Jul-99
Jul-00
Jul-01
Jul-02
Jan-98
Jan-99
Jan-00
Jan-01
Jan-02
Mar-98
Nov-98
Mar-99
Nov-99
Mar-00
Nov-00
Mar-01
Nov-01
Mar-02
May-98
Sep-98
May-99
Sep-99
May-00
Sep-00
May-01
Sep-01
May-02
25
Exchange Rate Depreciation and Inflation in 2001
and Jan.-Aug./2002
30
24.1
25
20.9
20
15
10.4
10
7.7
7.6
6.5
4.8
3.7
5
0
Exchange Rate Depreciation
Administered Prices Inflation
Inflation Rate (IPCA)
Market Prices Inflation
2001
2002 (up to August)
26
Percentage of the Total and Percentage Variation
in the Period
Contributions to Inflation in 2001
Inertia
10
Market Prices
Inflation Excluding
Exchange Rate Pass-
0.7
Through and Inertia
2.4
28
2.9
Exchange Rate Pass-
Through
38
1.7
Administered Prices
Inflation Excluding
Exchange Rate Pass-
Through and Inertia
24
27
Percentage of the Total and Percentage Variation
in the Period
Contributions to Inflation in Jan-Aug/2002
Inertia
Market Prices Inflation
17
Excluding Exchange Rate
Pass-Through and Inertia
0.8
22
1.1
Administered Prices
Inflation Excluding
0.6
Exchange Rate Pass-
2.3
Through and Inertia
Exchange Rate Pass-
Through
13
48
28
VAR estimation Impulse Response Functions

• Output, adm. prices, market prices, EMBI,
  exchange rate, interest rate

Response of ADMP to ER
Response of FREEP to ER
Response of ER to ER
1.2
6
0.5
5
1.0
0.4
4
0.8
3
0.3
0.6
2
0.2
1
0.4
0
0.1
0.2
-1
0.0
0.0
-2
-0.2
-3
-0.1
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
29
Pass-Through from Exchange Rate
Consumer Price
Model Estimation
Administered Prices
Market Prices
Index (IPCA)
Structural Model and Priori
Information
25
12
16
VAR Estimation
20
8
14
12
5
8
Simple AS Curve
21
9 (IT period)
30
Methodology to calculate effects of shocks to
administered prices
31
Strategy of Central Bank

• Central Bank accomodates primary effects of
  shocks to administered prices but not
  second-round effects
• To calculate the primary effect of the shocks to
  administered prices, the effects of the
  exchange-rate change and inflation inertia are
  excluded

32
Strategy of Central Bank

• and only part of inertia is neutralized in the
  current year
• Therefore, the components concerning the primary
  effect of shocks to administered prices and part
  of the inflation inertia are excluded from the
  expected inflation. The new value is then
  compared to the inflation target.

33
Example of Calculation of Inflation Excluding
Inertia Effect from Previous Year and the
Primary Effect of the Shock to Administered Prices
Item

(a) Inflation forecast for year t
5.0
(b) Contribution of year t-1 inertia to market
price inflation in year t
0.6
(c) Inertia from year t-1 to be neutralized in
year t 1 ( b/2)
0.3
(d) Administered price inflation forecast for
year t
9.0
1.2
(f) Inertia effects of year t-1 on year t
administered price inflation
0.2
(g) Exchange rate impact on administered price
inflation
0.5
(h) Primary effect of the shock to administered
prices ( e - f -g)
0.5
(i) Inflation forecast excluding the inertia
effect and the primary effect of the shocks to
4.2
administered prices ( a - c - h)
34
Institutional Design of Inflation Targeting

• Core
• Escape clauses
• Tolerance intervals
• Establishment of targets

35
Headline versus core measure

• Core by exclusion two problems
• - agents are concerned about the whole basket of
  consumption
• - In Brazil, in the past, there were some
  exclusions in the official price index that would
  affect the credibility if a core by exclusion
  were adopted
• Only Canada and Thailand adopt it. Australia, New
  Zealand, and Czech Republic have abandoned it
  (see Ferreira and Petrassi, 2002)

36
Inflation Rate and (trimmed mean) Core Inflation
Inflation Rate and Core Inflation
199601 - 200208 ( p.m.)
2.0
1.5
1.0
0.5
0.0
-0.5
Jul-96
Jul-97
Jul-98
Jul-99
Jul-00
Jul-01
Jul-02
Jan-96
Apr-96
Oct-96
Jan-97
Apr-97
Oct-97
Jan-98
Apr-98
Oct-98
Jan-99
Apr-99
Oct-99
Jan-00
Apr-00
Oct-00
Jan-01
Apr-01
Oct-01
Jan-02
Apr-02
Inflation Rate
Core Inflation

37
Trimmed mean core good predictor of trend
inflation
Granger Causality Test Core Inflation and
Inflation Rate (IPCA)
Sample 199601 - 200206
1 lag
2 lags
Null Hypothesis
F-statistic
P-value
F-statistic
P-value
Core Inflation does not Granger Cause Inflation
Rate
4.83
0.0311
7.14
0.0015
Inflation Rate does not Granger Cause Core
Inflation
0.04
0.8479
0.98
0.3797
38
Escape clauses

• the circumstances under which central banks can
  justify non-fulfillment of targets are set in
  advance (natural disasters, change in
  international prices, in indirect taxes, etc.)
  New Zealand, South Africa, Czech Republic,
  Switzerland
• the adoption of escape clauses could signal
  central bank would be lenient in the context of a
  credibility under construction

39
Tolerance intervals (bands)

• Reasons
• limits of forecasting models
• unexpected shocks
• lags in the effects of monetary policy
• Brazil short period of stabilization, frequent
  and large shocks
• - tolerance intervals enlarged from 2 p.p. to
  2.5 p.p. above and below the central target

40
Establishment of targets

• Brazil target is set in June of year t for t2
• Change in the target for 2003 from 3.25 to 4.0
  
• Loss of credibility of change versus loss of
  credibility of pursuing a unlikely target

41
Main conclusions

• The inflation-targeting regime in Brazil is
  relatively new, but has shown to be important to
  achieve low levels of inflation rate even in a
  context of large shocks
• Regime has faced many challenges
• 1. construction of credibility
• - CB has reacted strongly to inflation
  expectations
• - inflation expectations under control
• - reduction in the degree of inflation
  persistence

42
Main conclusions

• 2. change in relative prices -gt inflation
• - CB has developed methodology to estimate
  primary - effects of administered prices
• 3. exchange rate volatility -gt inflation
• - Institutional design
• - CB has enlarged the band size