German financial state aid during Covid-19 pandemic: Higher impact among digitalized self-employed

Figures & data

Figure 1. Conceptual framework.

Source: Own depiction

Table 1. Digitalization level among German self-employed persons in 2020.

	products and services	internal processes	customer relations/ distribution channels
before COVID-19 crisis
low	.38	.24	.33
medium	.16	.21	.22
high	.29	.38	.26
during COVID-19 crisis
decreased	.07	.07	.09
unchanged	.54	.63	.53
increased	.22	.12	.19

Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Figure 2. Percentage of households with fast broadband access ((≥ 1 Gbits/s) by federal state.

Source: FMDT (2020).

Figure 3. Common support.

Source: Own calculations.

Table 2. ATT by digitalization level before and during the crisis.

	main	digitalization level pre-crisis			digitalization level during crisis
		low	medium	high	decreased	unchanged	increased
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.065	.053	.070	.110	.076	.074	.036
SE	(.023)	(.033)	(.109)	(.040)	(.091)	(.035)	(.045)
p-value	.004	.113	.521	.006	.401	.037	.427
N matched	6,284	3,210	798	2,344	729	3,403	2,258
N unmatched	472	53	121	230	89	23	145
N total	6,756	3,263	919	2,574	818	3,426	2,403

Notes: Epanechnikov-kernel estimator with min/max trimming criterion. Standard errors were bootstrapped with B=1,999 replications. Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table 3. ATT for ventures with low digitalization level by digitalization area.

	pre-crisis low digitalization level in …
	products/services	internal processes	customer relations/distribution channels
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.014	.043	.004
SE	(.035)	(.045)	(.036)
p-value	.684	.346	.902
N matched	3,036	1,782	2,539
N unmatched	310	191	281
N total	3,346	1,973	2,820

Notes: Epanechnikov-kernel estimator with min/max trimming criterion. Standard errors were bootstrapped with B=1,999 replications. Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table 4. ATT for ventures with high digitalization level by digitalization area.

	pre-crisis high digitalization level in …
	products/services	internal processes	customer relations/distribution channels
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.108	.088	.139
SE	(.045)	(.038)	(.045)
p-value	.016	.020	.002
N matched	1,790	2,893	1,856
N unmatched	179	85	159
N total	1,969	2,978	2,015

Notes: Epanechnikov-kernel estimator with min/max trimming criterion. Standard errors were bootstrapped with B=1,999 replications. Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table 5. ATT by region.

	Berlin, Hamburg, and Bremen	other regions
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.123	.052
SE	(.061)	(.025)
p-value	.044	.038
N matched	1,626	4,937
N unmatched	133	60
N out of common support	0	0
N total	1,759	4,997

Notes: Epanechnikov-kernel estimator with min/max trimming criterion. Standard errors were bootstrapped with B=1,999 replications. Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table 6. ATT with nearest neighbor-matching.

	main	digitalization level pre-crisis			digitalization level during crisis
		low	medium	high	decreased	unchanged	Increased
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.068	.047	.080	.107	.094	.084	.016
SE	(.022)	(.029)	(.089)	(.028)	(.054)	(.025)	(.038)
p-value	.002	.111	.366	.000	.083	.001	.671
N total	6,756	3,263	898	2,574	759	3,426	2,403

Notes: Nearest neighbor matching estimator with two neighbors and replacement. Observations with perfect treatment probability were discarded. Robust standard errors were estimated following Abadie and Imbens (2016). Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table A1. Self-employed persons by region.

Region	share inhabitants in German population	self-employed persons by region
		Germany	Sample
South
Baden-Württemberg	.13	.13	.10
Bavaria	.16	.17	.17
West
North-Rhine Westphalia	.22	.19	.22
Hesse	.08	.08	.08
Saarland	.01	.01	.01
Rhineland Palatinate	.05	.05	.04
North
Bremen	.01	.01	.01
Hamburg	.02	.03	.05
Lower Saxony	.10	.09	.08
Schleswig-Holstein	.04	.04	.03
East
Mecklenburg-Western Pomerania	.02	.02	.01
Brandenburg	.03	.03	.03
Saxony	.05	.05	.05
Saxony-Anhalt	.03	.02	.01
Thuringia	.03	.03	.02
Berlin	.04	.07	.11
Sum	1.00	1.00	1.00

Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), Destatis 2022, IfM Bonn (2019), own calculations.

Table A2. ATT for ventures with low digitalization level by value chain component (NN2 matching).

	pre-crisis low digitalization level in …
	products/services	internal processes	customer relations/distribution channels
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.013	.057	.020
SE	(.035)	(.031)	(.031)
p-value	.709	.072	.513
N total	3,346	1,973	2,820

Notes: Nearest neighbor matching estimator with two neighbors and replacement. Observations with perfect treatment probability were discarded. Robust standard errors were estimated following Abadie and Imbens (2016). Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Table A3. ATT for ventures with high digitalization level by value chain component (NN2 matching).

	pre-crisis high digitalization level in …
	products/services	internal processes	customer relations/distribution channels
Dependent variable: self-assessment to remain self-employed within the next 12 months (yes/no)
ATT	.088	.050	.142
SE	(.059)	(.027)	(.033)
p-value	.135	.066	.000
N total	1,958	2,978	2,015

Notes: Nearest neighbor matching estimator with two neighbors and replacement. Observations with perfect treatment probability were discarded. Robust standard errors were estimated following Abadie and Imbens (2016). Source: Survey on self-employment during COVID-19 crisis in Germany - wave 2020 (doi:10.25652/diw_data_S0021.1), own calculations.

Federal Ministry for Digital and Transport (FMDT). 2020. Aktuelle Breitbandverfügbarkeit in Deutschland (Stand Mitte 2020). Berlin.

 Google Scholar

Abadie, A., and G. Imbens. 2016. “Matching on the Estimated Propensity Score.” Econometrica 84 (2): 781–807. doi:10.3982/ECTA11293.

 Web of Science ®Google Scholar