2965
Views
81
Downloads |
Governance quality vs. stimulus size: fiscal policy effectiveness during the COVID-19 pandemic
Mert Topcu* 
Mert Topcu
Affiliation: Alanya Alaaddin Keykubat University, Faculty of Economics, Administrative and Social Sciences, Department of Economics, Antalya, Türkiye
 0000-0001-8236-9810
Correspondence
mert.topcu@alanya.edu.tr
Mustafa Alpin Gulsen *
Mustafa Alpin Gulsen
Affiliation: Alanya Alaaddin Keykubat University, Faculty of Economics, Administrative and Social Sciences, Department of Economics, Antalya, Türkiye
0000-0002-2860-4469
Article | Year: 2025 | Pages: 251 - 272 | Volume: 49 | Issue: 2 Received: December 25, 2024 | Accepted: March 23, 2025 | Published online: June 7, 2025
|
FULL ARTICLE
FIGURES & DATA
REFERENCES
CROSSMARK POLICY
METRICS
LICENCING
PDF
|
Principle component
|
Proportion of variance
|
Cumulative proportion
|
|
PC1
|
85.1
|
85.1
|
|
PC2
|
6.7
|
91.9
|
|
PC3
|
4.5
|
96.5
|
|
PC4
|
2.0
|
98.6
|
|
PC5
|
0.8
|
99.4
|
|
PC6
|
0.5
|
100.0
|
Source: Authors’ calculation.
|
Variables
|
Obs.
|
Mean
|
Std. dev.
|
Min.
|
Max.
|
|
Growth
|
144
|
-0.010
|
0.053
|
-0.179
|
0.165
|
|
Capital
|
144
|
7.116
|
1.541
|
2.978
|
10.561
|
|
Education
|
143
|
2.079
|
0.457
|
0.470
|
2.653
|
|
Lockdown policy
|
144
|
3.988
|
0.451
|
1.868
|
4.497
|
|
Fiscal stimulus
|
144
|
2.293
|
1.211
|
-1.647
|
8.202
|
|
Governance quality
|
144
|
-0.150
|
1.379
|
-2.304
|
2.222
|
|
|
Growth
|
Capital
|
Education
|
Lockdown policy
|
Fiscal stimulus
|
|
Growth
|
1.000
|
|
|
|
|
|
Capital
|
0.187
|
1.000
|
|
|
|
|
Education
|
0.191
|
0.690
|
1.000
|
|
|
|
Lockdown policy
|
-0.163
|
0.416
|
0.435
|
1.000
|
|
|
Fiscal stimulus
|
0.048
|
0.436
|
0.376
|
0.227
|
1.000
|
|
|
Panel A: Full Sample
|
Panel B: Robustness Check
|
|
Regimes
|
Coefficients
|
Coefficients
|
|
Regime 1
|
-0.045***
(0.009)
|
-0.027**
(0.014)
|
-0.040***
(0.012)
|
-0.041***
(0.015)
|
|
Regime 2
|
0.017
(0.015)
|
0.001
(0.013)
|
0.038
(0.018)
|
0.002
(0.016)
|
|
Regime 3
|
-
|
0.039**
(0.019)
|
-
|
0.039**
(0.019)
|
|
Threshold variable
|
fiscal stimulus
|
δ
|
fiscal stimulus
|
δ
|
|
Threshold order
|
1: 1.395
|
1: -0.736
2: 0.145
|
1: 1.798
|
1: -0.761
2: 0.145
|
Note: Regressions include a regime varying constant. The results of regime-varying control variables are not reported. δ is an interaction of governance quality with the reverse of the fiscal stimulus. White-corrected standard errors in parentheses. *** and ** indicate significance at 1% and 5% levels, respectively. The maximum number of thresholds has been set to 2. Source: Authors’ calculation.
|
Below threshold value 1
|
Above threshold value 2
|
|
Angola
|
Gabon
|
Myanmar
|
Algeria
|
Denmark
|
Israel
|
Saudi Arabia
|
|
Bangladesh
|
Haiti
|
Nicaragua
|
Australia
|
Estonia
|
Jamaica
|
Slovak Republic
|
|
Belarus
|
Iraq
|
Nigeria
|
Austria
|
Finland
|
Malaysia
|
South Korea
|
|
Central African Republic
|
Kenya
|
Republic of Congo
|
Belgium
|
France
|
Mauritius
|
Spain
|
|
Chad
|
Laos
|
Tajikistan
|
Botswana
|
Georgia
|
Namibia
|
Sweden
|
|
Democratic Congo
|
Liberia
|
Tanzania
|
Canada
|
Germany
|
Netherlands
|
Switzerland
|
|
Ecuador
|
Libya
|
Turkmenistan
|
Chile
|
Hong Kong
|
New Zealand
|
United Arab Emirates
|
|
Egypt
|
Malawi
|
Uzbekistan
|
Costa Rica
|
Hungary
|
Norway
|
United Kingdom
|
|
Eswatini
|
Mali
|
Zimbabwe
|
Czech Republic
|
Ireland
|
Portugal
|
Uruguay
|
Source: Authors’ calculation.
|
Albania
|
Burundi
|
Egypt
|
Hong Kong
|
Latvia
|
Nepal
|
Romania
|
Tanzania
|
|
Algeria
|
Cambodia
|
El Salvador
|
Hungary
|
Lesotho
|
Netherlands
|
Russia
|
Thailand
|
|
Angola
|
Cameroon
|
Eritrea
|
India
|
Liberia
|
New Zealand
|
Rwanda
|
Togo
|
|
Argentina
|
Canada
|
Estonia
|
Indonesia
|
Libya
|
Nicaragua
|
Saudi Arabia
|
Trinidad Tobago
|
|
Australia
|
Central African Republic
|
Eswatini
|
Iran
|
Lithuania
|
Niger
|
Senegal
|
Tunisia
|
|
Austria
|
Chad
|
Ethiopia
|
Iraq
|
Madagascar
|
Nigeria
|
Serbia
|
Turkey
|
|
Azerbaijan
|
Chile
|
Finland
|
Ireland
|
Malawi
|
Norway
|
Sierra Leone
|
Turkmenistan
|
|
Bahrain
|
China
|
France
|
Israel
|
Malaysia
|
Oman
|
Singapore
|
UAE
|
|
Bangladesh
|
Colombia
|
Gabon
|
Italy
|
Mali
|
Pakistan
|
Slovak Republic
|
Uganda
|
|
Belarus
|
Costa Rica
|
Gambia
|
Jamaica
|
Mauritania
|
Panama
|
Slovenia
|
UK
|
|
Belgium
|
Cote Ivory
|
Georgia
|
Japan
|
Mauritius
|
Papua New Guinea
|
South Africa
|
Ukraine
|
|
Benin
|
Croatia
|
Germany
|
Jordan
|
Mexico
|
Paraguay
|
South Korea
|
United States
|
|
Bolivia
|
Cyprus
|
Ghana
|
Kazakhstan
|
Moldova
|
Peru
|
Spain
|
Uruguay
|
|
Bosnia and Herzegovina
|
Czech Republic
|
Greece
|
Kenya
|
Mongolia
|
Philippines
|
Sri Lanka
|
Uzbekistan
|
|
Botswana
|
Democratic Republic of Congo
|
Guatemala
|
Kosovo
|
Morocco
|
Poland
|
Sudan
|
Vietnam
|
|
Brazil
|
Denmark
|
Guinea
|
Kuwait
|
Mozambique
|
Portugal
|
Sweden
|
Yemen
|
|
Bulgaria
|
Dominican Republic
|
Haiti
|
Kyrgyz Republic
|
Myanmar
|
Qatar
|
Switzerland
|
Zambia
|
|
Burkina Faso
|
Ecuador
|
Honduras
|
Laos
|
Namibia
|
Republic of Congo
|
Tajikistan
|
Zimbabwe
|
Table 1PCA results (in %) DISPLAY Table
Table 2Summary statistics DISPLAY Table
Table 3Correlation matrix DISPLAY Table
Table 4Estimation results DISPLAY Table
Table 5Country list by threshold level DISPLAY Table
Table A1Country coverage DISPLAY Table
* The authors would like to thank Co-Editor Katarina Ott and two anonymous reviewers for their valuable comments.
1 Note that real per capita income (2015 U.S.$ in constant prices) growth was -3,85 in 2020 while it was 5,47 in 2021 (World Bank WDI Database, 2024).
2 Notice that there is well-documented literature on the potential macroeconomic costs of fiscal packages provided in the pandemic era (see, i.e., Banerjee et al.,  2022; Dean, 2022; De Soyres, Santacreu and Young, 2022; Horton and El-Ganainy, 2022; among others).
3 According to Blanchard ( 2020), governments should be prepared to take action, but should not commit to a precise amount of fiscal expansion before they exactly estimate its impact on demand.
4 Because good governance stimulates the institutions expected to play a crucial role in facilitating economic growth (Knack and Keefer, 1995; 1997; Acemoglu, Johnson and Robinson, 2001; Acemoglu and Robinson, 2012; among others), there is a strong connection – even interchangeability – between governance quality and institutional quality.
5 Gregory ( 2022), the only study that attempts to address this issue in the context of environmental and social governance with micro-level data, finds that non-financial firms that manage environmental and governance risk better were able to perform better over the pandemic.
6 Countries in the sample are available in appendix.
7 While this index is widely used in the literature and provides a standardized measure for cross-country comparisons (Gros, Ounnas and Yeung, 2021; Ma et al., 2021), it has some limitations. Specifically, the index may not fully capture informal lockdowns, localized restrictions, or differences in enforcement levels across regions within a country. Moreover, reporting inconsistencies could exist among countries, particularly in cases where subnational governments implemented independent restrictions that were not reflected in national-level data. Despite these limitations, the index remains an effective proxy for lockdown stringency, given its broad coverage and methodological consistency across countries.
8 We use the version of the dataset released on May 7, 2021.
9 Governance quality is a multidimensional concept, and its impact on economic outcomes is often the result of the combined influence of various institutional factors. By using a single index derived from PCA, we focus on the aggregate effect of governance rather than the isolated impacts of each indicator. This approach allows us to analyse governance as a unified construct and prevents overemphasis on specific dimensions while maintaining methodological coherence.
10 To avoid negative values, the inverse hyperbolic sine transformation is applied where required.
11 As an additional methodological robustness check, it would be helpful to address potential endogeneity concerns related to governance quality and fiscal stimulus influenced by unobserved country characteristics. While panel data approaches could help mitigate such concerns, our analysis based on cross-sectional data limits is limited in its ability to control for unobserved heterogeneity.
|
|
June, 2025
II/2025
|
