The joint effects of formal and informal institutions on entrepreneurial startups: a global perspective

Figures & data

Figure 1. Flowchart of “TEA” production.

Table 1. Description of model variables.

Measure	Definition	Possible value	Source
Dependent variable
Entrepreneurial startups	If the respondents are someone who actively involved in starting a new firm (nascent entrepreneur) or owing and managing an existing business less than 3.5 years (young business owner)	0 – No	GEM-APS
		1 – Yes
Independent variables
Resource-based index	The resource-based factor includes three opportunities in relation to the financial, human, and social capitals	Range from 0 to 3	GEM-APS
Regulatory quality	The regulatory quality is developed by the World Bank by aggregating individual variables from different data sources.	Range from −2.5 to +2.5	WGI
Uncertainty avoidance	Uncertainty avoidance index (UAI) is collected from Hofstede’s well-known study of cultural dimensions across countries.	Range from 0 to 1	HCI
Control variables
Gender	What is your gender?	1 – Female	GEM-APS
		2 – Male
Age/age squared	What is your current age (in years)?		GEM-APS
Education attainment	What is the highest qualification you have achieved?	1 – Primary or below
		2 – Secondary
		3 – Post-secondary
		4 – Graduate experience
GDP per capital	Gross domestic product (GDP) at purchasing power parity (PPP) per capita		WDI
Population	The natural logarithm of a country’s total population		WDI

GEM-APS: Global Entrepreneurship Monitor – Adult Population Survey; WDI: World Development Indicators; WGI: Worldwide Governance Indicators; HCI: Hofestede Cultural Index.

Table 2. Correlation matrix.

	Mean	S.D.	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Entrepreneurial startups (1)	0.12	0.32	1.000
Age (2)	39.50	11.45	0.077**	1.000
Gender (3)	0.60	0.48	0.065**	0.003	1.000
Education attainment (4)	2.25	0.94	0.016**	0.035**	0.086**	1.000
GDP per capital (5)	2.34	0.47	0.007	0.024**	0.010*	0.064**	1.000
Total population (6)	1.11	1.58	0.098**	0.011*	0.010*	0.043**	0.410**	1.000
Resource-based index (7)	1.18	1.46	0.160**	0.054**	0.033**	0.065**	−0.121**	0.152**	1.000
Regulatory quality (8)	0.65	0.84	0.126**	0.158**	0.035**	−0.207**	0.312**	−0.139**	−0.043**	1.000
Uncertainty avoidance (9)	43.27	22.33	−0.109**	−0.147**	−0.019**	0.141**	−0.088**	0.090**	0.030**	−0.654**	1.000

Note: **p < 0.01;*p < 0.05.

Table 3. Multilevel logistic regression analysis results.

	Model 1		Model 2		Model 3
	Coefficient	S.E.	Coefficient	S.E.	Coefficient	S.E.
Fixed effects parameters
Individual-level control variables
Age			0.010	(0.009)	0.009	(0.009)
Age-squared/100			−0.027*	(0.011)	−0.028*	(0.011)
Gender			0.402***	(0.034)	0.418***	(0.034)
Educational attainment			0.050**	(0.017)	0.051**	(0.017)
Country-level control variables
GDP per capital			0.687*	(0.277)	0.299	(0.299)
Total population			0.153*	(0.077)	0.164*	(0.073)
Resource-based index			0.264***	(0.011)
Country-level predictor
Regulatory Quality (RQ)					0.336*	(0.141)
Random effects parameters
sigma_μ	0.81		0.62		0.59
% of variance, ρ	19.77		15.87		15.21
Model Fit
Number of Observations	41,156		41,156		41,156
Number of Countries	46		46		46
Log-likelihood	−13470.6		−13053.6		−13341.6
Akaike Information Criterion (AIC)	26945.2		26125.2		26701.1
Bayesian Information Criterion (BIC)	26962.5		26202.8		26778.8
Likelihood ratio test	–		*** (vs model 1)		*** (vs model 2)

Note: ***p < 0.001; **p < 0.01;*p < 0.05; + p < 0.1.

Table 4. Multilevel logistic regression analysis results.

	Model 4		Model 5		Model 6
	Coefficient	S.E.	Coefficient	S.E.	Coefficient	S.E.
Fixed effects parameters
Individual-level control variables
Age	0.010	(0.009)	0.009	(0.009)	0.009	(0.009)
Age-squared/100	−0.027*	(0.011)	−0.028*	(0.011)	−0.027*	(0.011)
Gender	0.404***	(0.034)	0.418***	(0.034)	0.404***	(0.034)
Educational attainment	0.030+	(0.017)	0.051**	(0.017)	0.032+	(0.018)
Country-level control variables
GDP per capital	0.341	(0.288)	0.552*	(0.281)	0.385	(0.306)
Total population	0.141*	(0.070)	0.179*	(0.076)	0.126+	(0.074)
Resource-based index (Res)	0.282***	(0.012)			0.341***	(0.025)
Country-level predictor
Regulatory Quality (RQ)	0.255^+	(0.138)			0.369	(0.311)
Uncertainty avoidance (UA)			−0.718	(0.509)	−0.760	(0.793)
Cross-level interaction terms
RES*RQ	0.044***	(0.012)			0.102***	(0.026)
RES*UA					−0.230**	(0.079)
RQ*UA					−0.185	(0.619)
RES*RQ*UA					−0.184**	(0.058)
Random effects parameters
sigma_μ	0.54		0.63		0.54
% of variance, ρ	14.11		16.08		14.11
Model Fit
Number of Observations	41,156		41,156		41,156
Number of Countries	46		46		46
Log-likelihood	−13044.1		−13343.2		−13038.0
Akaike Information Criterion (AIC)	26110.1		26704.5		26106.0
Bayesian Information Criterion (BIC)	26205.0		26782.1		26235.4
Likelihood ratio test	*** (vs model 4)		*** (vs model 3)		*** (vs model 6)

Note: ***p < 0.001; **p < 0.01;*p < 0.05; +p < 0.1.

Table 5. Multilevel logistic regression analysis results.

	Model 7		Model 8
	Coefficient	S.E.	Coefficient	S.E.
Fixed effects parameters
Individual-level control variables
Age	0.011	(0.011)	−0.005	(0.017)
Age-squared/100	−0.022	(0.014)	−0.021	(0.020)
Gender	0.311***	(0.041)	0.596***	(0.062)
Educational attainment	0.014	(0.021)	0.082*	(0.032)
Country-level control variables
GDP per capital	0.831*	(0.324)	−0.938	(0.582)
Total population	0.170*	(0.084)	0.020	(0.130)
Individual-level predictors
Resource-based Index	0.811***	(0.040)	0.302***	(0.040)
Country-level predictor
Regulatory Quality (RQ)	0.547	(0.664)	0.247	(0.167)
Cross-level two-way interaction
RES*RQ	0.193***	(0.054)	0.025+	(0.014)
Random effects parameters
sigma_μ	0.49		0.38
% of variance, ρ	12.97		10.36
Model Fit
Number of Observations	21,062		20,094
Number of Countries	26		20
Log-likelihood	−8435.4		−4580.1
Akaike Information Criterion (AIC)	16892.9		9182.2
Bayesian Information Criterion (BIC)	16980.4		9269.2
Likelihood ratio test	*** (vs model 4)		*** (vs model 4)

Note: ***p < 0.001; **p < 0.01;*p < 0.05; + p < 0.1.

Figure 2. Interaction between alertness to opportunities and the regulatory quality.

Figure 3. Interaction between alertness to opportunities and the regulatory quality in low uncertainty avoidance regime.

Figure 4. Interaction between fear of failure and the regulatory quality in high uncertainty avoidance regime.

Table

Representative sources	Variables
EIU	Unfair competitive practices
	Price controls
	Discriminatory tariffs
	Excessive protections
	Discriminatory taxes
GCS	Burden of government regulations
	Prevalence of non-tariff barriers
GWP	Confidence in judicial system
	Confidence in the police force
	Have you been assaulted or mugged?
	Have you had money property stolen from you or another household member?
HER	Investment freedom
	Financial freedom
IPD	Ease of starting a business governed by local law?
	Ease of setting up a subsidiary for a foreign firm?
	Share of administered prices
	Does the State subsidize commodity prices (i.e. food and other essential goods, excluding oil)?
	Efficiency of competition regulation in the market sector (excluding financial sector)
PRS	Investment profile
WMO	Regulatory burden. The risk that normal business operations become more costly due to the regulatory environment. This includes regulatory compliance and bureaucratic inefficiency and/or opacity. Regulatory burdens vary across sectors so scoring should give greater weight to sectors contributing the most to the economy.
	Tax inconsistency. Tax inconsistency also captures the risk that fines and penalties will be levied for noncompliance with a tax code that appears disproportionate or manipulated for political ends.
Non-representative Sources	Variables
ADB	Regional integration
	Trade policy
	Business regulatory environment
ASD	Trade policy
	Business regulatory environment
BPS	How problematic are labor regulations for the growth of your business?
	How problematic are tax regulations for the growth of your business?
	How problematic are customs and trade regulations for the growth of your business?
BTI	Market organization
IFD	Enabling conditions for rural financial services development
	Investment climate for rural businesses
	Access to agricultural input and product markets
	Trade policy
PIA	Business regulatory environment
	Trade policy
WCY	Protectionism does not impair the conduct of your business
	Competition legislation is efficient in preventing unfair competition
	Capital markets (foreign and domestic) are easily accessible
	The legal and regulatory framework encourages the competitiveness of enterprises
	Foreign investors are free to acquire control in domestic companies
	Public sector contracts are sufficiently open to foreign bidders
	Real personal taxes do not discourage people from working or seeking advancement
	Labor regulations (hiring/firing practices, minimum wages, etc.) do not hinder business activities
	Subsidies do not distort fair competition and economic development
WJP	Regulatory enforcement

ADB: African Development Bank Country Policy and Institutional Assessments; ASD: Asian Development Bank Country Policy and Institutional Assessments; BPS: Business Enterprise Environment Survey; BTI: Bertelsmann Transformation Index; CCR: Freedom House Countries at the Crossroads; EBR: European Bank for Reconstruction and Development Transition Report; EIU: Economist Intelligence Unit Riskwire & Democracy Index; GCB: Transparency International Global Corruption Barometer Survey; GCS: World Economic Forum Global Competitiveness Report; GWP: Gallup World Poll; HER: Heritage Foundation Index of Economic Freedom; HUM: Cingranelli Richards Human Rights Database and Political Terror Scale; IFD: IFAD Rural Sector Performance Assessments; IJT: iJET Country Security Risk Ratings; IPD: Institutional Profiles Database; IRP: IREEP African Electoral Index; MSI: International Research and Exchanges Board Media Sustainability Index; OBI: International Budget Project Open Budget Index; PIA: World Bank Country Policy and Institutional Assessments; PRC: Political Economic Risk Consultancy Corruption in Asia Survey; PRS: Political Risk Services International Country Risk Guide; RSF: Reporters Without Borders Press Freedom Index; VAB: Vanderbilt University Americas Barometer; WCY: Institute for Management and Development World Competitiveness Yearbook; WJP: World Justice Project Rule of Law Index; WMO: Global Insight Business Conditions and Risk Indicators.

Table

	Model 1		Model 2		Model 3
	Coefficient	S.E.	Coefficient	S.E.	Coefficient	S.E.
Fixed effects parameters
Individual-level control variables
Age			0.004	(0.006)	0.004	(0.006)
Age-squared/100			−0.029***	(0.008)	−0.029***	(0.008)
Gender			0.322***	(0.023)	0.327***	(0.023)
Educational attainment			0.013	(0.012)	0.019	(0.012)
Country-level control variables
GDP per capital			0.824*	(0.417)	0.855*	(0.414)
Total population			0.295**	(0.101)	0.287**	(0.101)
Resource-based index			0.207***	(0.012)
Country-level predictor
Regulatory Quality (RQ)					0.507***	(0.132)
Random effects parameters
sigma_μ	0.85		0.79		0.60
% of variance, ρ	20.54		19.37		15.43
Model Fit
Number of Observations	70,363		70,363		70,363
Number of Countries	48		48		48
Log-likelihood	−26476.7		−26030.2		−26179.9
Akaike Information Criterion (AIC)	52957.3		52078.4		52377.7
Bayesian Information Criterion (BIC)	52975.6		52160.9		52460.2
Likelihood ratio test	–		*** (vs model 1)		*** (vs model 2)

Note: ***p < 0.001; **p < 0.01; *p < 0.05; + p < 0.1.

Table

	Model 4
	Coefficient	S.E.
Fixed effects parameters
Individual-level control variables
Age	0.004	(0.006)
Age-squared/100	−0.029***	(0.008)
Gender	0.323***	(0.023)
Educational attainment	0.014	(0.012)
Country-level control variables
GDP per capital	0.850*	(0.406)
Total population	0.298**	(0.100)
Resource-based index (Res)	0.152***	(0.023)
Country-level predictor
Regulatory Quality (RQ)	0.448***	(0.132)
Cross-level interaction terms
RES*RQ	0.039**	(0.014)
Random effects parameters
sigma_μ	0.59
% of variance, ρ	17.95
Model Fit
Number of Observations	70,363
Number of Countries	48
Log-likelihood	−26019.9
Akaike Information Criterion (AIC)	52061.8
Bayesian Information Criterion (BIC)	52162.6
Likelihood ratio test	>*** (vs model 4)

Note: ***p < 0.001; **p < 0.01; *p < 0.05; + p < 0.1.