The influences of covid-19 pandemic on macroeconomic indexes for European countries

Figures & data

Table 1. The variables used for models

indicator code	indicator name
I8	Final consumption expenditure of general government
I14	Imports of goods and services
I19	Employers’ social contributions
I20	Subsidies
I21	Taxes on production and imports less subsidies
I26	Real labour productivity per person
I36	Wages and salaries Agriculture, forestry and fishing
I37	Wages and salaries Industry (except construction)
I44	Final consumption expenditure of households
I55	Government consolidated gross debt
School	Requiring school closing (all levels or only some levels)
Workplace	Require closing (or work from home) for all or for some categories
Stay_home	Require not leaving house
Internal	Internal movement restrictions
Ban_arrivals	Ban arrivals from some regions or all regions or total border closure
Cases	1 if number of cases in 2020Q1 is above 2000, and 1 if number of cases in 2020Q2 is above 10,000

Table 2. The dummy variable criteria

Variable	2020Q1	2020Q2
School	20	79
Workplace	15	67
Stay_home	8	32
Internal	7	32
Ban_arrivals	24	79
Cases	2000	10,000

Figure 1. Correlation matrix for datasets 2019-Q2 (left) and 2020-Q2 (right).

Figure 2. Correlation matrix – original variables and PC.

Table 3. The main results of the principal component analysis

Dataset	Number of variables	Number of PC	% of information	% of loss
2018Q1	10	5	80.33%	19.67%
2018Q2	10	5	81.30%	18.70%
2019Q1	10	5	79.25%	20.75%
2019Q2	10	5	82.63%	17.37%
2020Q1	10	5	82.64%	17.36%
2020Q2	10	5	87.36%	12.64%

5 PCs consider over 80% of total information (79.25% for 2019-Q1), as the lost information is 100%-% of information.

Figure 3. First PC score and GDP −2018 Q1 vs. 2019 Q1.

Figure 4. The number of adopted closure policies −2020 Q1 vs. Q2.

Output obtained in R software; Starting February-March 2020, the number of cases increased rapidly in most European countries, so one closure measure was not enough anymore.

Figure 5. Regression models results for 2018 and 2019 - Q1 and Q2.

Output obtained in R software²; t statistics is in parantheses,***, **, * refers to level of significance: 1%, 5%, 10%

Figure 6. Variance inflation factor, White test and Durbin Watson test for 2018 and 2019 datasets – Q1 and Q2.

Figure 7. Regression models results for 2020 - Q1.

Outputs obtained in R software, t statistic is in parentheses, ***, **, * refers to level of significance: 1%, 5%, 10%

Figure 8. Variance inflation factor, White and Durbin-Watson tests results for 2020-Q1.

Outputs obtained in R software; the Dummy_var $\in \left\{school,workplace,stay\mathrm{\_}home,internal,ban\mathrm{\_}arrivals,cases\right\}$ for each model (example: the name of the model 2020Q1D1 corresponds the first model in Figure 7, where the dummy variable is school)

Figure 9. Regression models results for 2020 – Q2.

Outputs obtained in R software, t statistic is in parentheses, ***, **, * refers to level of significance: 1%, 5%, 10%

Figure 10. Regression models results for 2020 – Q2.

Outputs obtained in R software, t statistic is in parentheses, ***, **, * refers to level of significance: 1%, 5%, 10%

Hlavac, M. 2018. “Stargazer: Well-Formatted Regression and Summary Statistics Tables.” R package version 5.2.2. https://CRAN.R-project.org/package=stargazer

 Google Scholar