Transcript Error Correction Models and Real Convergence – Case of

Nominal and Real Convergence
of Slovak Republic and Poland to Eurozone
Rudolf Gavliak, Vladimír Úradníček, Emília Zimková
The 5th Chorzow Conference of Banking and Finance
Knowledge and Funds Transfer into the Sectors of the EU Economy
September 26 – 27, 2008
Content
1.
2.
3.
4.
Introduction
Methodology and Data
Outcome of Analyses
Conclusion
Introduction
Classification of the Convergence
Structural
Nominal Convergence (1)
Price
stability
Public Finance
Exchange Rate
HICP
Def/HDP Debt/HDP
Interest
rates
10 y
2006
2,1
-2,7
29,4
No
4,8
3,8
2007
3,0
-1,6
28,7
No
2,0
4,3
2008
4,4
-1,4
28,1
No
8,4
4,5
Hungary 2006
4,0
-9,2
65,5
No
-6,5
7,1
2007
7,9
-5,5
66,0
No
4,9
6,7
2008
7,5
-4,0
66,5
No
-2,7
6,9
Czech
Rep.
Source: Convergence Report 2008
Nominal Convergence (2)
Price
stability
Public Finance
Exchange Rate
Interest
rates
HICP
Def/HDP Debt/HDP ERM II Apprec
/Deprec 10 y
Poland 2006
1,3
-3,8
47,6
No
3,2
5,2
2007
2,6
-2,0
45,2
No
2,9
5,5
2008
3,2
-2,5
44,5
No
6,3
5,7
Slovak 2006
Rep.
2007
4,3
-3,6
30,4
No
3,5
4,4
1,9
-2,2
29,4
No
9,3
4,5
2008
2,2
-2,7
35,5
No
2,5
4,5
Source: Convergence Report 2008
Structural Convergence
1.
The structural convergence is analysing convergence
of the economy according its sectors, employment,
inovations, research, economic reforms, social and
enviromental policies...
2.
It is also analysing the harmonisation of the economic
cycles and the synchronisation of economics shocks
(Gavliak, Úradníček, Zimková, 2007, SVAR
technique, the Blanchard-Quah teoretical approach).
Real Convergence
1.
The real convegence can be measured by economic
performance, the level of labour productivity, the level of
prices and wages (Barančok, 2007)
2.
Commonly used indicator to meassure the economic
convergence is the gross domestic product in the
purchasing power parity (HDPPPP).
Methodology
„Real economic convergence“ can be estimated by
 β convergence
 σ convergence
 Cointegration and Error Correction
Models
Methodology - β convergence
β convergence is a concept which is estimating the
speed of convergence of the individual country to the
average of clusstered countries.

 When the partial correlation between growth in
income over time and its initial level is negative, there
is β convergence.
Methodology - σ convergence
When the dispersion of real per capita income across
a group of economies falls over time, there is
σ convergence.
Methodology – Cointegration and ECM
 Long lasting mutual trend of analysed indicators is
searched by cointegration.
 In a short term there might be some deviations
which are studied by error correction models (ECM).
 The output is twofold: we analyze the long-lasting
equilibrium of the real economy and secondly we
estimate the duration whithin which the searched
variables are returning to the long-lasting equilibrium
in the case of the short-term deviations.
Methodology – Cointegration and ECM
 The cointegrating equations may have intercepts and
deterministic trends
 We provided tests for the following five possibilities
considered by Johansen
Johansen tests
1. Series y have no deterministic trends and the
cointegrating equations do not have intercepts
H 2  r  : Πyt 1  Bxt    yt 1.
2. Series y have no deterministic trends and the
cointegrating equations have intercepts
H1*  r  : Πyt 1  Bxt      yt 1  0 .
Johansen tests
3. Series y have linear trends but the cointegrating
equations have only intercepts:
H1  r  : Πyt 1  Bxt      yt 1  0    0 .
4. Both series y and the cointegrating equations have
linear trends:
H *  r  : Πyt 1  Bxt      yt 1  0  1t    0 .
Johansen tests
5. Series y have quadratic trends and the cointegrating
equations have linear trends:
H  r  : Πyt 1  Bxt      yt 1  0  1t      0   1t .
where  is the (non-unique) k x (k - r) matrix such

that    0 and rank    k.
 

Data and Software
 For analyses were used quarterly input data from
Eurostat datasource
 We analysed detrended data from first quarter 2000
till the fist quarter 2008
 EViews 4 software
Empirical Results
of Cointegration analyses – case of Slovakia
The cointegration equation, which is characterising the
long-lasting equilibrium, has the following
specification:
g SK = 3,87g EA,
where
g SK – y-on-y growth of the Gross Domestic Product in
Slovakia (in %),
g EA – y-on-y growth of the Gross Domestic Product in
Euroarea (in %).
Empirical Results
of Cointegration analyses – case of Slovakia
Error Correction Coefficients estimated in the Error
Correction Model
D(RASTHDP_SR)
-0.164680 (0.05673)
D(RASTHDP_EA12)
0.099963 (0.05393)
Empirical Results
of Cointegration analyses – case of Poland
The cointegration equation, which is characterising the
long-lasting equilibrium, has the following
specification:
g PL = 1,72g EA,
where
g PL – y-on-y growth of the Gross Domestic Product in
Poland (in %),
g EA – y-on-y growth of the Gross Domestic Product in
Euroarea (in %).
Empirical Results
of Cointegration analyses – case of Poland
Error Correction Coefficients estimated in the Error
Correction Model
D(RASTHDP_PL)
-0.1488 (0.06414)
D(RASTHDP_EA12)
0.132470 (0.06315)
Conclusion
The results of cointegration analyses proved solid the
real convergence of the Slovak Republic and Poland to
Euroarea.
The entry of both countries to the EMU might be a new
incentive to the dynamic economic development.
To speed up this proces the implementation of the
responsible economic, fiscal and labour market policies,
the introduction a new structure of the economy based on
the new technologies are requested.