Transcript RPM_Proyecciones

Use of Chilean business surveys
in conjunctural assessment and
short-term forecasting
Michael Pedersen
Central Bank of Chile
Fourth joint EC-OECD workshop on
business and consumer opinion surveys
BANCO CENTRAL DE CHILE
13 October 2009
The presentation
1. Short presentation of the Chilean business
surveys.
2. Case studies: Information in surveys about
Chilean economic activity.
1. Evaluating information in aggregated indicators.
2. Using surveys for short-term forecasting.
3. Information in cross-checking answers.
3. Final remarks.
2
The Chilean business surveys are designed
in line with the recommendations in the
OECD handbook
Some main characteristics:
 Frequency: monthly.
 Sample: fixed panel of the largest firms,
random selection of smaller ones.
 Total sample: 607 firms (16% of GDP).
 4 sectors: mining, manufacturing, retail and
construction.
 Presentation of results: diffusion indices
calculated with simple balances.
3
The series are available from November
2003
Difussion indices
90
90
80
80
70
70
60
50
60
50
40
30
40
30
20
20
2004
2005
IMCE
Source: ICARE / UAI
4
2006
ICOM
2007
ICIN
2008
ICOT
2009
ICMI
Three case studies are presented to
illustrate the information in the surveys
 Because of the very small sample, the use of
surveys in the conjunctural analysis has so
far mainly been on an ad-hoc basis.
 However, now almost six years of data are
available and it is possible to analyze the
statistical properties of the series – on a
preliminary basis.
5
1a. Relatively high coefficients of
correlation for manufacturing and retail
Diffusion indices and annual growth rates
Retail sector (26%)
Manufacturing sector (39%)
70
15
60
60
10
55
50
10
5
50
5
45
0
40
0
40
35
-5
-10
30
-5
30
-15
2004
2005
2006
ICOM (lhs)
2007
2004
2008
2005
ICIN (lhs)
Re tail se ctor (rhs)
Mining (17%)
15
80
10
70
5
60
50
0
-5
40
-10
30
-15
2005
2006
ICMI (lhs)
2007
2008
Mining se ctor (rhs)
Sources: ICARE / UAI, Pozo and Stanger (2009) and own calculations.
6
2007
2008
Industrial prod. (rhs)
Construction (18%)
90
2004
2006
15
80
20
70
15
60
10
50
5
40
0
30
-5
20
-10
2004
2005
2006
ICOT (lhs)
2007
2008
Construction (rhs)
1b. Tests suggest that surveys of
manufacturing and retail Granger cause the
activity indicators
Cross correlation coefficients
0,80
0,80
0,60
0,60
0,40
0,40
0,20
0,20
0,00
0,00
-0,20
-0,20
-6
-4
-2
0
2
4
6
ICOM / Re tail
ICIN / Ind. prod.
ICMI / Mining
ICOT / Construction
IMCE / GDP
Sources: ICARE / UAI, Pozo and Stanger (2009) and own calculations.
Note: Negative numbers on the first axis indicate that the business survey leads activity.
Tests for Granger causality
IMCE ICOM ICIN ICMI ICOT
Activity
Survey
survey 0.21
activity 0.10
0.95
0.00
0.62
0.00
0.87
0.76
0.26
0.43
Source: Own calculations.
Note: p-values for the null hypothesis of no Granger causality tested in bivariate VAR models with the number of lags selected according
to Schwarz information criteria.
7
2. Preliminary estimations indicate that the
general survey contains information which
is useful for short-term forecasting
Out-of-sample one-step-ahead forecasting exercise
a
RMSE
BS-model betterb
D-Mc
Total Retail Manufacturing sector Mining Construction
0.76 0.69
0.62
0.80
1.03
75% 67%
67%
42%
42%
0.00 0.07
0.00
0.10
0.59
Source: Own calculations.
Note: aRMSE of the business survey model divided by the RMSE of the AR model. bPercentage of the twelve observations where the
business survey model predicts better than the AR model. cp-value of the Diebold and Mariano (1995) test for the hypothesis that
the models have equal predictive power against the alternative that the business survey model is better.
8
3a. Why did the stock accumulation
fall in the last three quarters?
 Hypothesis 1: Because of restrictive financial
conditions, firms could not borrow money to
finance production.
 Hypothesis 2: Because of expectations of
lower sale, the optimal stock level implies a
reduction.
 Decompose the fraction of firms which
replied that stocks were higher than desired:
p( I tA )  p( I tA Vt A ) p(Vt A )  p( I tA Vt N ) p(Vt N )  p( I tA Vt B ) p(Vt B )
9
3b. Firms considered stock levels too high
because of their expectation of future
demand
Stock flow and cross-checking business survey answers
A: Stock flow / GDP (%)
3,0
B: Actual demand (%)
40
35
2,0
30
1,0
25
0,0
20
- 1,0
15
- 2,0
10
- 3,0
5
- 4,0
0
2004
- 5,0
96
40
98
00
02
04
06
08
C: Future production (3M) (%)
Low
40
35
35
30
30
25
25
20
20
15
15
10
10
5
5
0
2004
0
2004
2005
2006
Dow n
2007
Unchange d
2008
2009
Up
2005
2006
2007
Normal
2008
2009
High
D: Future general situation (6M) (%)
2005
Worse
2006
2007
Same
2008
2009
Be tte r
Source: Echavarría et al. (2009)
Note: Figure A shows the stock flows as a percentage of GDP. B-D show the proportion of the firms that considered that their stock levels were
higher than desired and also considered that: (B) actual demand was low, normal and high, respectively; (C) future production will go down, not
change and go up, respectively, and (D) the future general situation of the company will be worse, the same and better, respectively.
10
Final comments
 The preliminary results are promising with
respect to the contents in Chilean business
surveys about the economic activity.
 Several questions remain unanswered:
 Are surveys affected by seasonality?
 Do surveys lead annual growth rates, or rather
monthly?
 Levels or changes in surveys serve as leading
indicators?
 Ongoing research in the Central Bank of
11
Chile aims at answering some of these
questions.
Use of Chilean business surveys
in conjunctural assessment and
short-term forecasting
Michael Pedersen
Central Bank of Chile
Fourth joint EC-OECD workshop on
business and consumer opinion surveys
BANCO CENTRAL DE CHILE
13 October 2009