POSIBILITAŢI DE PERFECŢIONARE A ACTIVITAŢII DE …

Download Report

Transcript POSIBILITAŢI DE PERFECŢIONARE A ACTIVITAŢII DE …

Academy of Economic Studies
Doctoral School of Finance and Banking
Economic Growth, Fiscal Size and Volatility: A
Panel Assessment for EU Developing Economies
MSc Student: Dan Matei
Supervisor: PhD Professor Moisa Altar
Dissertation paper outline
 Introduction
Literature review
Methodology
Empirical Analysis
- Data
- Results and Discussion
- Robustness Analysis
Conclusions
Bibliography
Introduction

In the EU area, the gradual loss of monetary policy as an instrument to offset country-specific
disturbances naturally places the onus on fiscal policy. European countries would thus be
facing a difficult trade-off between maintaining large governments to ensure sufficient
automatic fiscal stabilization and leaner ones to ensure efficiency and growth (there could be a
tension between the ‘Maastricht’ and the ‘Lisbon’ goals).
Debate between the need to ensure adequate macroeconomic stabilization and the reduction in the
size of governments that often accompanied efforts to boost market efficiency and promote
long-term growth
The importance of high-quality fiscal policies for economic growth, a firm control and, where
appropriate, reduction in public spending have been brought to the forefront by a number of
developments over the past decades
With a view to understand how to limit government size and restrict fiscal policy volatility, it is
quite relevant to assess which components of general government spending and revenue (both
in terms of size and volatility) have a negative effect on growth. Although the effect of
government expenditure volatility has been widely analyzed, the effect of volatility in the
components of public spending and revenue has not so far been widely addressed in the
literature
By regressing economic growth on budgetary items and on a set of other relevant variables we
evaluate whether the allocation of taxes and public expenditures has been useful to promote
growth in a panel of European countries for the period 1996-2007. The outcome of the paper
suggests that for several components of general government revenue and spending both size
and volatility measures have a negative effect on growth, and that restrictions on these
variables should be pursued.
Literature review

Wagner’s Law- the long-run tendency for government spending as a share of some national income
aggregate such as GDP to grow in the course of economic development has become more or less a
stylized fact in public finance

Keynesian perspective- public expenditure should act as a stabilizing force and move in a
countercyclical direction

Barro (1990) constitutes one of the first attempts at endogenizing the relationship between growth
and fiscal policies, distinguishing four categories of public finances: productive vs. non-productive
expenditures and distortionary vs. non-distortionary taxation

Levine-Renelt (1992) found that most results from earlier studies on the relationship between longrun growth and fiscal policy indicators are fragile to small changes in the conditioning set

Easterly-Rebello (1993) public transportation, communication and educational investment are
positively correlated with growth per capita and aggregate public investment is negatively correlated
with growth per capita

Poot (1999) in a survey of published articles in 1983-1998 did not find conclusive evidence for the
relationship between government consumption and growth, still found empirical support for the
negative effect of taxes on growth and reported definitive results on the positive link between growth
and education spending

Afonso and Tanzi (2005) finds that a well-defined institutional framework and ‘high quality’ public
finances are important to support the long-run growth. Studying the efficiency of the public sectors of
23 industrialized OECD countries, they noted that countries with large public sectors show more
equal income distribution, while countries with small public sectors report significantly higher
indicators than countries with medium-sized or big public sectors.
Literature review (contd.)

Fiscal volatility - There is little consensus on the sign of the effects of government expenditure
volatility on growth, restrictions on government expenditure volatility may have both positive and
negative effects on long-run growth. A crucial variable to determine the sign of these effects is
business-cycle volatility

Ramey and Ramey (1995) find a negative relation between the business-cycle volatility and growth
in cross-country data and this relation is robust to various controls

Aghion (2005) find that the effect of volatility on growth survives when one controls for the level of
financial development, leaving open the possibility that volatility has a causal effect on growth (the
negative relation between volatility on growth tends to be stronger in countries with lower financial
development )

On the mechanism through which fiscal policy can affect business cycles, Lane (2003) shows that
restrictions on government expenditure, and thus lower government expenditure volatility, result in a
slower adjustment of the economy to unexpected shocks

In contrast, Fatas and Mihov (2003) present evidence that aggressive use of discretionary fiscal
policy generates undesirable output volatility and leads to lower growth. Not only discretionary
changes but also transitory (and cyclical) changes in fiscal policy may increase output volatility and
thereby reduce output growth

Ayagari, Christiano and Eichenbaum (1992) temporary changes in fiscal policy may have a
significant impact on interest rate volatility and this, in turn, will reduce long-run growth. Furceri
(2007b) analyzing a panel of 99 countries from 1970-2000, shows that a 1 percent increase in
government expenditure business cycle volatility determines a decrease of 0.78 percentage points in
the long-run rate of growth

The survey of different empirical studies shows that an objective and unambiguous overall
catalogue of “high quality”-expenditure items is not feasible. There is no cookbook for growth.
Economics gives an idea of the major ingredients, but it does not clearly tell the recipe.
Methodology
 The inclusion of particular control variables in a growth regression can
wipe out the negative bivariate relationship between growth and the measure
of government size (Easterly and Rebelo, 1993)
 Levine and Renelt (1992) found that robust cross-country growth correlates
to the average investment share of GDP, the initial log of GDP per capita,
initial human capital and the average growth rate of the population
 Initial income is often used to test the convergence hypothesis
 Opening to trade is beneficial to economic growth on average, allows the
dissemination of knowledge and technological progress, still the aftermath of
trade openness varies considerably across countries and depends on a variety
of conditions related to the structure of the economy and its institutions
 Output volatility: tends to have negative effects on long-term economic
growth, welfare, and income inequality, particularly in developing countries.
As main justifications for short-run “stabilization” policies (policies aimed at
reducing volatility, The World Bank and the IMF routinely advise
governments to reduce fluctuations to achieve higher growth rates
Methodology (contd.)
 Time span- cross-country growth regressions make use of large time spans (30- 40 years) and
consider the average value of growth determinants over this time period. As argued by Afonso
and Furceri (2008), this could raise problems such as endogeneity and significant simultaneity.
Cross-section analysis over long time spans may fail to capture growth causality effects of
taxation
 The analysis is focused on combined cross-section time-series regressions using three four-year
periods from 1996 to 2007, and we use pooled country and fixed effects
 The model- two growth equations respectively for general government revenue and expenditure:
gi,t = α1 + β1Ri,t +γ1 R2i,t +δ1σRi,t +φ1Xi,t +εi,t (1)
gi,t = α2 + β2Ei,t +γ2 E2i,t +δ2σEi,t +φ2Xi,t +εi,t (2)
where
the index i (i=1, …, 10) denotes the country, the index t (t= 1996-1999, 2000-2003, 2004-2007) indicates the period,
α1 and α2 stand for the individual effects to be estimated for each country i.
g is the growth rate of real GDP per capita,
R is the vector of general government revenue variables as percentage of GDP,
E is the vector of general government expenditure variables as percentage of GDP,
σR is the vector of revenue volatility variables, and σE is the vector of expenditure volatility variables,
X is a vector of control variables (initial level of output per capita, output volatility, investment share, population
growth and openness).
Both regressions also include square terms for R and E with a view to test the possible effect on economic growth of
different government sizes.
Empirical analysis- Data
 Sources of data are European Commission AMECO (Annual Macro- Economic
Data), supplemented by EUROSTAT database, covering the period 1995-2007
 The panel consists in 10 EU members and emerging economies: Bulgaria,
Czech Rep, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, Slovenia and
Slovak Rep
Variable
Growth
(g)
rate
Initial output
Population
growth
Investment
share of GDP
Openness
Output
volatility
Fiscal
variables
Definition of the variable
Abbreviation
The four year average in the growth rate of
GDP
per
head
of
population
(PPS
EU25=100)- AMECO
The log of real GDP per head of population at
the beginning of each time period - AMECO
The average of the annual log difference of
total population AMECO
Share of total economy investment in real
GDP- AMECO
Share of exports and imports of goods and
services at 2000 prices in real GDP- AMECO
Standard deviation in each period of the
cyclical component of real GDP- AMECO
Total Revenue (R), Direct Taxes (RD),
Indirect Taxes (RI), Social Contributions
(RC), Total Expenditure (EX), Government
Consumption (EY), Government Investment
(EI), Transfers (ET), Subsidies (ES)-current
prices as in ratios to current GDP
GS
GDP
POP
INV
OPEN
VOL
R, E
Empirical analysis- Data (contd.)
 Advantages
 homogeneity - all 10 EU countries are emerging market economies
 data quality and cross-country comparability are likely to be of a good standard for the
EU members (fiscal variable in ESA 95)
 Drawbacks
 fiscal data availability for the studied economies is rather limited
 only 3 observations per country for each variable, employing 4 year growth periods
 Two measures for both government revenues and expenditures (the aggregates and
components): the relative share of each variable as a percentage of GDP and the
volatility of the cyclical component for each fiscal variable
 For volatility measures, all fiscal variables are converted into constant prices using the
GDP deflator. To compute the cyclical component for each fiscal variable, Hodrick and
Prescott Filter was set with the smoothness parameter (λ) equal to to 6.25. In this way, as
pointed out by Ravn and Uhlig (2002), the Hodrick-Prescott filter produces cyclical
components comparable to those obtained by the Band-Pass filter.
 The analysis excludes those fiscal variables that have a residual importance on the
public budget or whose interpretation is not clear
Empirical analysis- Data (contd.)
15
10
5
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
RO
SI
2006
2007
-5
-10
-15
BG
CZ
EE
LV
LT
HU
PL
SK
Table B . Total public revenue and expenditure as of % in GDP
Revenues
Expendit
Revenues
Expendit
Revenues
Expendit
Bulgaria
1996-1999
41.89
47.45
2000-2003
42.99
45.12
2004-2007
40.70
38.29
Czech R
Estonia
Hungary
Latvia
Lithuania
Poland
Romania
Slovenia
Slovak R
38.84
38.61
45.75
38.06
36.79
42.11
46.10
43.73
41.76
39.28
35.92
42.77
33.40
33.49
38.57
37.53
44.20
37.63
41.33
36.19
42.96
36.41
33.19
39.09
33.06
44.00
34.74
42.80
39.19
51.91
38.77
42.07
46.13
49.96
45.99
48.93
44.98
35.45
48.55
35.58
35.98
43.43
40.40
47.45
45.04
44.01
33.61
50.21
36.82
34.12
43.05
34.84
45.26
37.48
Revenues
Expendit
Change in pp
-1.19
-9.16
2.49
-2.42
-2.79
-1.65
-3.60
-3.01
-13.04
0.27
-7.02
1.21
-5.59
-1.70
-1.95
-7.95
-3.08
-15.11
-0.74
-11.45
Results and Discussion
Autor (s)
Data period and
coverage
43 developing
countries, yearly
data, 1970-1990
22 OECD
countries, yearly
data, 1970-1995
Estimation method /
model
Fixed-effects (Five-year
forward moving
average dep. Variable)
Fixed-effects, random
effects (Five-year
averages)
Bassanini and
Scarpetta (2001)
21 OECD
countries, 19711998
Pooled Mean Group
Estimator
Folster and
Henrekson
(2001)
Bose, Haque
and Osborn
(2003)
23 OECD
countries, 19701995
30 developing
countries, decade
averages, 19701990
Fixed-country and
period effects (five-year
averages)
OLS (Decade average
dep var.)
Romero de Avila
and
Strauch (2007)
15 European
countries, 19602001
Long-term coefficients
estimated by variables
in levels
Afonso and
Furceri (2008)
EU 15 and OECD
Fixed-country and
period effects (five-year
averages)
Devarajan,
Swaroop and
Zou (1996)
Kneller,
Bleaney and
Gemmell (1999)
Main results
Excess public capital expenditure
for their data set.
Negative effect distortionary
taxation Negative impact non
productive expenditures (social
transfers) Negative effect deficit
Positive impact of public
investment Unclear effect of
public current expenditure.
Negative impact of taxation
Significant negative effect for
total government spending;
negative effect of total taxes.
Identify the importance of
education and government
spending for economic growth in
their set of countries. Also find a
significant correlation with capital
expenditure.
Negative impact of total
expenditure on growth.Positive
impact of direct taxation, indirect
taxation and public investment.
Negative effect of government
consumption, transfers, and social
security revenues.
Negative impact of total revenue
and expenditure (size and
volatility) on growth.
Total general gov revenue and Growth
Total general gov expenditure and Growth
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/01/08 Time: 12:05
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/01/08 Time: 12:01
Sample: 1 3
Likelihood Ratio Test (LR)
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Variable
Coefficient
Std. Error
t-Statistic
Prob.
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?EX
?EV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
72.35072
-7.452652
0.033865
0.287571
-0.645998
-3.420679
-0.206032
-35.90981
26.15442
2.940737
0.018252
0.106261
1.010027
20.46026
0.104064
12.73928
2.766290
-2.534280
1.855477
2.706262
-0.639585
-0.167186
-1.979866
-2.818825
0.0160
0.0249
0.0863
0.0180
0.5336
0.8698
0.0693
0.0145
C
79.62294
24.66883
3.227674
0.0066
?GDP
-8.576761
2.715306
-3.158672
0.0075
?OPEN
0.043079
0.015015
2.869154
0.0132
?INV
0.479516
0.086206
5.562420
0.0001
?POP
0.101703
0.873380
0.116448
0.9091
?VOL
-2.015172
17.03405
-0.118303
0.9076
?RT
-0.252000
0.103690
-2.430307
0.0303
?RV
-53.13048
14.92202
-3.560543
0.0035
Fixed Effects (Cross)
BG_--C
2.891983
CZ_--C
-2.231262
EE_--C
1.752868
LV_--C
3.411685
LT_--C
1.977317
HU_--C
-3.365424
PL_--C
-0.976839
RO_--C
0.862814
SI_--C
-0.740042
SK_--C
-3.583100
Effects Specification
0.312628
-0.891112
0.331291
2.287829
1.400373
-1.315973
-0.813290
-1.118800
0.911902
-1.104848
Effects Specification
Cross-section fixed (dummy variables)
Cross-section fixed (dummy variables)
R-squared
0.941045
Mean dependent var
2.965056
Adjusted R-squared
0.868486
S.D. dependent var
2.614630
S.E. of regression
0.948193
Akaike info criterion
3.028569
Sum squared resid
11.68792
Schwarz criterion
3.822581
F-statistic
12.96926
Prob(F-statistic)
0.000017
Log likelihood
Durbin-Watson stat
-28.42853
2.105775
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.921397
0.824654
1.094860
15.58333
-32.74322
2.247730
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.316215
4.110227
9.524194
0.000099
Government revenue composition and Growth Government expenditure composition and Growth
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/02/08 Time: 14:45
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/01/08 Time: 12:14
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Likelihood Ratio Test (LR)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?INV
?POP
?OPEN
?VOL
?RC
?RD
?RI
?RCV
?RDV
?RIV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
92.10316
-8.140363
0.226942
0.372617
0.029658
14.19782
-0.869911
-0.654178
-0.100484
-92.07420
-29.72830
-10.83928
22.21132
2.362437
0.096309
0.785986
0.013590
18.43361
0.424020
0.223529
0.347201
22.04242
15.88484
20.94634
4.146677
-3.445748
2.356399
0.474075
2.182393
0.770213
-2.051580
-2.926589
-0.289411
-4.177137
-1.871489
-0.517479
0.0025
0.0073
0.0429
0.6467
0.0570
0.4609
0.0704
0.0169
0.7788
0.0024
0.0941
0.6173
2.331263
-0.441158
1.477878
1.318199
-1.305672
-2.573520
-0.630894
-0.274016
1.136243
-1.038323
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?INV
?OPEN
?POP
?VOL
?EI
?ES
?ET
?EY
?EIV
?ESV
?ETV
?EYV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
104.7285
-9.630068
0.134199
0.035115
-2.029919
-17.68561
0.280551
-1.495720
-0.373082
-0.603592
-9.096458
-18.69154
-19.74383
31.24314
27.79607
3.112627
0.168179
0.020162
1.148612
20.89954
0.299895
0.920913
0.417786
0.242555
3.031519
7.465719
6.184621
22.41553
3.767745
-3.093871
0.797953
1.741643
-1.767280
-0.846220
0.935500
-1.624170
-0.892999
-2.488475
-3.000627
-2.503650
-3.192407
1.393817
0.0070
0.0175
0.4511
0.1251
0.1205
0.4254
0.3807
0.1484
0.4015
0.0417
0.0199
0.0408
0.0152
0.2060
Effects Specification
Effects Specification
Cross-section fixed (dummy variables)
Cross-section fixed (dummy variables)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.972638
0.911834
0.776354
5.424534
-16.91418
2.178442
1.978706
-1.374290
-0.261079
2.585113
-1.164567
-3.316032
-0.628892
-3.486577
5.340207
0.327410
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
2.527612
3.508450
15.99631
0.000092
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.975975
0.900467
0.824887
4.763072
-14.96359
2.439251
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
2.530906
3.605157
12.92544
0.000975
Total general revenue Size and Growth
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/09/08 Time: 00:20
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Total general expenditure Size and Growth
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/08/08 Time: 16:59
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Likelihood Ratio Test (LR)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?RT^2
?RV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
72.36091
-8.324898
0.042109
0.477871
0.156597
-3.133775
-0.003057
-53.35587
23.69199
2.729112
0.015231
0.087510
0.884737
17.19895
0.001322
15.14364
3.054235
-3.050406
2.764765
5.460738
0.176998
-0.182207
-2.312323
-3.523320
0.0092
0.0093
0.0161
0.0001
0.8622
0.8582
0.0378
0.0037
C
?GDP
?OPEN
?INV
?POP
?VOL
?EX^2
?EV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
64.94703
-7.107881
0.032386
0.287045
-0.582702
-3.999283
-0.002335
-36.32117
25.33219
2.949936
0.018796
0.107623
1.016925
20.65448
0.001235
12.87477
2.563814
-2.409504
1.723011
2.667127
-0.573004
-0.193628
-1.890842
-2.821113
0.0236
0.0315
0.1086
0.0194
0.5764
0.8495
0.0811
0.0144
2.932885
-2.356132
1.798650
3.425212
2.099944
-3.371045
-1.040173
1.035649
-0.958997
-3.565992
Effects Specification
0.425986
-1.069837
0.469644
2.278679
1.503218
-1.243772
-0.927904
-0.936891
0.595419
-1.094542
Effects Specification
Cross-section fixed (dummy variables)
Cross-section fixed (dummy variables)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.939247
0.864475
0.962544
12.04439
-28.87918
2.076776
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.058612
3.852624
12.56137
0.000021
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.919762
0.821008
1.106182
15.90729
-33.05186
2.222071
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.336791
4.130803
9.313679
0.000112
Robustness analysis
 The inclusion of country specific effects has the advantage of controlling for unobserved
country heterogeneity, it could lead to misleading conclusion in the analysis of the results
 Re-estimating the growth equations excluding country dummies, the results remain robust
to the change
 To control for a possible endogeneity problem in the regression, the equations were reestimated using the initial level of government spending and revenue-to- GDP ratios
Table C. Robustness control
HP 6.25
HP 100
Difference
-35.909**
-23.922**
-13.327*
(-2.81)
(-2.51)
(-1.58)
Average volatility
0.030
0.060
0.068
Effect
-1.073
-1.435
-0.906
Notes: t-statistics are in parenthesis.
*, **, *** - Statistically significant at the 10, 5 and 1 percent level respectively.
Total general gov revenue and Growth
Total general gov expenditure and Growth
including only period dummies
including only period dummies
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/08/08 Time: 17:23
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/08/08 Time: 17:22
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Likelihood Ratio Test (LR)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?RT
?RV
Fixed Effects (Period)
1--C
2--C
3--C
37.92442
-3.036599
0.007881
0.207064
-2.655265
-22.85483
-0.286801
-26.93313
13.25092
1.523758
0.010071
0.112629
0.875854
24.14660
0.096315
21.53513
2.862021
-1.992836
0.782532
1.838455
-3.031629
-0.946503
-2.977731
-1.250660
0.0096
0.0601
0.4431
0.0809
0.0066
0.3552
0.0074
0.2255
-0.377782
-0.365811
0.743594
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?EX
?EV
Fixed Effects (Period)
1--C
2--C
3--C
43.07583
-3.580702
0.016002
0.195564
-1.624522
-16.18842
-0.273875
-28.22993
9.965472
1.150954
0.007742
0.080012
0.716700
17.00960
0.054512
11.48521
4.322508
-3.111073
2.066900
2.444184
-2.266669
-0.951722
-5.024116
-2.457937
0.0003
0.0055
0.0519
0.0239
0.0347
0.3526
0.0001
0.0232
0.190278
-0.122914
-0.067364
Effects Specification
Effects Specification
Period fixed (dummy variables)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.767977
0.663567
1.516561
45.99913
-48.97953
1.673357
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.931969
4.399035
7.355372
0.000108
Period fixed (dummy variables)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.866716
0.806738
1.149434
26.42396
-40.66426
1.818559
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.377618
3.844683
14.45056
0.000001
Total general gov revenue and Growth
initial share
Total general gov expenditure and Growth
initial share
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/09/08 Time: 01:46
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Dependent Variable: ?GS
Method: Pooled Least Squares
Date: 07/09/08 Time: 01:31
Sample: 1 3
Included observations: 3
Cross-sections included: 10
Total pool (balanced) observations: 30
Likelihood Ratio Test (LR)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?RT0
?RV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
84.97852
-9.322428
0.043165
0.488989
0.103089
-4.715752
-0.201735
-70.17708
32.86627
3.458115
0.016551
0.099072
0.966134
19.05833
0.125387
18.62933
2.585584
-2.695812
2.608043
4.935688
0.106702
-0.247438
-1.608905
-3.767021
0.0226
0.0183
0.0217
0.0003
0.9167
0.8084
0.1316
0.0024
3.535717
-2.477439
1.744033
3.411158
2.065010
-3.792842
-1.275051
1.594174
-1.123011
-3.681749
Variable
Coefficient
Std. Error
t-Statistic
Prob.
C
?GDP
?OPEN
?INV
?POP
?VOL
?EX0
?EV
Fixed Effects (Cross)
BG_--C
CZ_--C
EE_--C
LV_--C
LT_--C
HU_--C
PL_--C
RO_--C
SI_--C
SK_--C
65.24022
-7.147506
0.036034
0.319227
-0.601574
-12.39607
-0.115351
-38.66577
26.81962
3.085459
0.019274
0.109559
1.064831
20.51973
0.076828
13.56090
2.432556
-2.316513
1.869524
2.913749
-0.564948
-0.604105
-1.501428
-2.851269
0.0302
0.0375
0.0842
0.0121
0.5817
0.5562
0.1571
0.0136
Effects Specification
Effects Specification
Cross-section fixed (dummy variables)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
0.928498
0.840495
1.044235
14.17554
-31.32296
Durbin-Watson stat
2.146637
0.716744
-1.495695
0.810049
2.743530
1.857939
-2.462766
-0.902785
0.039254
-0.256152
-1.050119
Cross-section fixed (dummy variables)
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
2.965056
2.614630
3.221531
4.015543
10.55074
Prob(F-statistic)
0.000056
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
0.912814
0.805508
1.153084
17.28484
-34.29764
2.119379
Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
2.965056
2.614630
3.419843
4.213855
8.506656
0.000184
Conclusion
 The overall results suggest that both fiscal size and volatility tend to
hamper growth in EU developing economies
 A percentage point increase in the share of total revenue
(expenditure) would reduce output growth by 0.25 and 0.21 percentage
points respectively, for the EU developing countries
 Among total revenue the variables that are most detrimental to
growth, both in terms of size and volatility, are direct taxes and social
contributions
 Among government outlays, subsidies and government consumption
have a significantly negative impact on growth, government
investment and transfers does not significantly affect growth
 In terms of volatility, the government transfers and public subsidies
volatility have the largest negative effect on growth in the sample, in
addition the investment volatility have a negative and statistically
significant effect on growth in the EU developing countries
Conclusion (contd.)
 Restrain in government consumption and subsidies enhances
economic growth, on the revenue-side, contributions to social security
and direct taxation seem to be an obstacle for higher growth
 The result of this analysis should be taken with some prudence and the
estimated elasticities have to be analyzed with concern
 Insightful results for policy makers when deciding which components
of public finances to adjust
 The national policies appear to be a complex package and future
researchers may wish to focus on interactions and synergies among
fiscal policies as opposed to the influence of any particular variable
 The analysis can be improved in several ways:
 the channels through which the composition of the public budget
affects economic growth may be addressed in a specific context;
 one could investigate the optimal size and the nature of the relationship
between the role of the various components of government spending
and revenue and growth;
 the decomposition of public expenditure may be extended to include
transfers between the different levels of government and transfers from
supranational levels of government (European Commission)
References














Afonso, A. and D. Furceri (2008), “Government size, composition, volatility and economic growth”, European
Central Bank Working Paper no. 849.
Afonso, A., L. Schuknecht, and V. Tanzi (2005), “Public Sector Efficiency: An International Comparison”, Public
Choice, 123 (3-4), 321-347.
Afonso, A., W. Ebert, L. Schuknecht, and M. Thöne (2005), “Quality of public finances and growth,” European
Central Bank Working Paper n. 438.
Aghion, P., G. Angeletos, A. Banerjee, and K. Manova (2005), “Volatility and Growth: Credit Constraints and
Productivity-Enhancing Investment”, NBER Working Papers, 11349.
Aghion, P. and I. Marinescu (2007), “Cyclical budgetary policy and economic growth: What do we learn from OECD
panel data?”.
Ayagari, S., L. Christiano, and M. Eichenbaum (1992), “The Output, Employment, and Interest Rate of Government
Consumption”, Journal of Monetary Economics, 30, 73-86.
Barro, R. and X. Sala-i-Martin (1995), Economic Growth. McGraw-Hill, New York.
Bird, R. (1971), “‘Wagner’s law: a pooled time-series, cross section comparisons,” National Tax Journal, 38, 209-218.
Bose, N., M. Haque, and D. Osborn (2003), "Public Expenditure and Economic Growth: A Disaggregated Analysis for
Developing Countries", Centre for Growth and Business Cycle Research Discussion Paper Series (University of
Manchester), No. 30.
Cavallo, A. (2007), “Output volatility and openness to trade: A reassessment”, Inter-American Development Bank, Working paper n.604.
de Ávila, D. and R. Strauch (2003), “Public finances and long-term growth in Europe evidence from panel data analysis” ECB Working
Paper N. 246.
Easterly, W. and S. Rebelo (1993), “Fiscal Policy and Economic Growth: An Empirical Investigation,” Journal of Monetary Economics 32
(3), 417-458.
Fatás, A. (2002), “The Effects of Business Cycles on Growth” in Loayza, N. and R. Soto, (eds.) Economic Growth: Sources, Trends and
Cycles, Central Bank of Chile.
Fatás, A. and I. Mihov (2003), “The Case for Restricting Fiscal Policy Discretion”, Quarterly Journal of Economics,
118, 1419-1447
References (contd.)














Fölster, S. and M. Henrekson (2002), “Growth effects of government expenditures and taxation in rich countries”,
European Economic Review 45 (8), 1501-1520.
Furceri, D. (2007b), “Is Government Expenditure Volatility Harmful for Growth? A Cross-Country Analysis,” Fiscal
Studies 28 (1), 103-120.
Hemming, R., M. Kell, and S. Mahfouz (2002), “The Effectiveness of Fiscal Policy in Stimulating Economic Activity
- A Review of the Literature,” IMF Working Paper No.02/208.
Hodrick, R. and E. Prescott, (1997), “Postwar U.S. Business Cycles: An Empirical Investigations”, Journal of Money,
Credit and Banking, 29(1), 1-16.
Kneller, R., M. Bleaney, and N. Gemmell (1999), "Fiscal policy and growth: evidence from OECD countries", Journal
of Public Economics 74, 171-190.
Lane, P. (2003), “The Cyclical Behavior of Fiscal Policy: Evidence from the OECD”, Journal of Public Economics,
87, 2261-2275.
Levine, R. and D. Renelt (1991), “Cross-Country Studies of Growth and Policy”, Macroeconomic Adjustment and
Growth Division, Country Economics Department World Bank Working Paper no. 608.
Levine, R. and D. Renelt (1992), “A Sensitivity Analysis of Cross-Country Growth Regressions”, American Economic
Review, 82, 942-963.
Poot, J. 1999, "A meta-analytic study of the role of government in long-run economic growth," ERSA conference
papers, European Regional Science Association
Ramey, G. and V. Ramey (1995), “Cross-country evidence on the link-between volatility and growth”, American
Economic Review, 85, 1138-1151.
Ravn, M. and H. Uhlig (2002), “On Adjusting the Hodrick-Prescott Filter for the Frequency of Observations”, Review
of Economics and Statistics, 84, 371-380.
Rodriguez, F. and D. Rodrick (1998), “Trade policy and economic growth: A skeptics guide to the cross-national
evidence”, Journal of Political Economy, 106(5), 997-1032.
Sala-i-Martin, X. (1997), “I Just Ran Two Million Regressions”, AEA Papers and Proceedings 87, 178-183.
Scarpetta, S., A. Bassanini (2001), “Economic growth in the OECD area: Recent trends at the aggregate and sectoral
level”, OECD Working Papers no. 248.