**4.1. Descriptive statistics**

**Table 1** presents the descriptive and summary statistics for our measures. Some variables proved to be highly correlated, and therefore we also conducted a diagnostic test of multicollinearity (examining the variance inflation factors (VIFs) of all variables in the analyses and found that it was not likely to be a problem in this dataset. The VIFs were lower than 2.6 which is far from 5. A value of 5 indicates that a problem of multicollinearity may arise [90].

#### **4.2. Test of hypotheses**

In **Table 2**, we present the results of linear regressions with feasible generalized least squares (FGLS). Model 1 includes all the countries considered in the sample, Models 2–6 include interactions between informal institutions and the income of the counties. The Wald Chi square tests suggest that all the models are significant (p < 0.001) and have high explanatory power, explaining well over 60% of the variance of leadership. As expected, all informal factors are related to leadership behavior.

Hypothesis 1 suggests that the level of tolerance has a positive and significant effect on leadership behavior. Findings support Hypothesis 1. Hypothesis 2 suggests that social capital has a significant and positive influence on leadership. Our results support Hypothesis 2. Creativity shows a positive and significant relationship with leadership, supporting Hypothesis 3. On the other hand, as we expected, power value has a negative impact on leadership, supporting Hypothesis 4, and finally responsibility has a significant and positive influence on leadership behavior, supporting Hypothesis 5. Control variables such as low education have a significant negative impact on the dependent variable. Many studies have demonstrated that education made a difference in leadership [40, 86]. Finally, GDP-PPP has a positive significant impact on leadership. Scholars have typically argued that economic factors play a causal role in personal behavior [91]. Labor force participation has a positive effect on the dependent variable. On the other hand, the control of corruption shows counterintuitive results.

Hypothesis 6 proposed that the level of a country's development positively moderates the relationships of informal institutions with leadership behavior. Model 2 showed the interaction effect between development and tolerance. The coefficient was positive and statistically significant for tolerance, social capital, creativity and responsibility as we expected, and the coefficient of power was negative and significant. Although the main effect of tolerance was positive, the interaction of tolerance and level of development on leadership was negative and statistically significant (p < 0.001). The interaction terms show that while the level of development decreases,

**Variable** 1. Leadership

2. Tolerance 3. Social capital

4. Creativity

5. Power 6. Responsibility

7. Education 8. LnGDPPPP 9. Labor force 10. Control of corruption

\*

p < 0.1.

\*\*<sup>p</sup> < 0.05.

\*\*\*p < 0.01.

**Table 1.**

Descriptive statistic and correlation matrix.

0.379

0.996

0.213\*

0.311\*\*

0.293\*\*

0.032

−0.399\*\*\*

−0.484\*\*\*

−0.351\*\*\*

0.576\*\*\*

0.341\*\*\*

Informal Institutions and Leadership Behavior: A Cross-Country Analysis

http://dx.doi.org/10.5772/intechopen.75840

49

0.527 0.277 9.634 68.252

10.099

0.246\*\*

0.015

0.261\*\*

−0.163

−0.320\*\*\*

−0.098

−0.196\*

0.119

0.864

0.259\*

0.301\*

−0.067\*

−0.239\*

−0.595\*\*\*

−0.427\*\*\*

−0.502\*\*\*

0.187

−0.171

−0.172

−0.063

0.278\*

0.318\*\*\*

0.086

0.225

−0.200\*\*\*

−0.377\*\*\*

−0.056

−0.051

0.439\*\*\*

0.054 4.232 2.963

0.671

−0.337\*\*\*

−0.359\*\*\*

−0.124

0.396\*\*\*

0.392

0.361\*\*\*

0.181

0.115

0.046

0.275\*\*

0.214\*

**Mean**

7.136 0.706

0.128

0.388\*\*\*

0.721

**Std. dev.**

**1**

**2**

**3**

**4**

**5**

**6**

**7**

**8**

**9**


and fixed-effects models and we used the Hausman specification test [*X*<sup>2</sup>

ing with feasible generalized least squares (FGLS).

**4. Results**

48 Leadership

**4.1. Descriptive statistics**

**4.2. Test of hypotheses**

related to leadership behavior.

*X*<sup>2</sup> = 0.0003] in order to verify the choice of the fixed- or random-effects model. The test suggested the use of the fixed-effects specification. We have corrected heteroskedasticity, estimat-

**Table 1** presents the descriptive and summary statistics for our measures. Some variables proved to be highly correlated, and therefore we also conducted a diagnostic test of multicollinearity (examining the variance inflation factors (VIFs) of all variables in the analyses and found that it was not likely to be a problem in this dataset. The VIFs were lower than 2.6 which

In **Table 2**, we present the results of linear regressions with feasible generalized least squares (FGLS). Model 1 includes all the countries considered in the sample, Models 2–6 include interactions between informal institutions and the income of the counties. The Wald Chi square tests suggest that all the models are significant (p < 0.001) and have high explanatory power, explaining well over 60% of the variance of leadership. As expected, all informal factors are

Hypothesis 1 suggests that the level of tolerance has a positive and significant effect on leadership behavior. Findings support Hypothesis 1. Hypothesis 2 suggests that social capital has a significant and positive influence on leadership. Our results support Hypothesis 2. Creativity shows a positive and significant relationship with leadership, supporting Hypothesis 3. On the other hand, as we expected, power value has a negative impact on leadership, supporting Hypothesis 4, and finally responsibility has a significant and positive influence on leadership behavior, supporting Hypothesis 5. Control variables such as low education have a significant negative impact on the dependent variable. Many studies have demonstrated that education made a difference in leadership [40, 86]. Finally, GDP-PPP has a positive significant impact on leadership. Scholars have typically argued that economic factors play a causal role in personal behavior [91]. Labor force participation has a positive effect on the dependent variable. On the

Hypothesis 6 proposed that the level of a country's development positively moderates the relationships of informal institutions with leadership behavior. Model 2 showed the interaction effect between development and tolerance. The coefficient was positive and statistically significant for tolerance, social capital, creativity and responsibility as we expected, and the coefficient of power was negative and significant. Although the main effect of tolerance was positive, the interaction of tolerance and level of development on leadership was negative and statistically significant (p < 0.001). The interaction terms show that while the level of development decreases,

other hand, the control of corruption shows counterintuitive results.

is far from 5. A value of 5 indicates that a problem of multicollinearity may arise [90].

(7) = 30.73, Prob >

**Table 1.** Descriptive statistic and correlation matrix.


**Table 2.** Linear regressions with feasible generalized least squares (FGLS). leadership behavior is more sensitive to informal institutions such as tolerance. In other words, this negative interaction term indicates that the relationship between tolerance and leadership is stronger when there is a lower, rather than higher, level of development. This is in line with leadership literature that emphasizes the importance of tolerance and stress management in

Although not fully synonymous, the intolerance of ambiguity (an individual cognitive state) and uncertainty avoidance (a behavioral phenomenon) are concepts that are likely to be posi

tively related [92]. In Model 3, we presented a model with the interaction terms between social capital and development. In this model, the coefficients are positive and statistically signifi

cant for tolerance, creativity and responsibility. Otherwise, they are negative and statistically significant for power and not significant for social capital. The interaction of social capital and development with leadership was not significant. In Model 4, tolerance, social capital and responsibility were positive and significant and power was negative and significant. Contrary to expectations, the direct effect and interaction term for creativity and the level of develop

ment is not significant. Similar results have been found in studies such as Dubinsky et al. [93]. They argue that contradictory results could be explained because creativity intelligence seems vague or unformulated. Similarly, Model 5 presented the interaction term between power and development. Although the main effect of power was negative and significant, the interaction of power and development on leadership was not significant. Finally, in Model 6, we can see the interaction of responsibility and development with leadership. Although the main effect of responsibility was negative and significant, the interaction effect of power was posi

tive and significant. This interaction means that countries with high levels of development experienced a stronger positive impact of responsibility on leadership. As we can see, all models confirm the importance of informal institutions, especially when these institutions are

leadership. Our results do not support Hypothesis 6, since not all informal institutions were

The current research integrates insights from the leadership literature and proposes institu

tional economics (particularly sociocultural approach) as a fresh perspective to advance lead

ership research, especially when we link leadership with the construction of an individual and collective identity. This study contributes to a better understanding of the mechanisms through which informal institutions such as cultural values, attitudes and practices influ

ence leadership. Although informal factors such as tolerance, creativity, social capital, power and responsibility were found to be important predictors of the decision to be a leader, their effects are somewhat complex. Our results demonstrated that the level of country develop

ment exercise a complex pattern of the effects on the relationship between informal institu

tions and leadership behavior. This is one of the few studies that aim to integrate the study of

*R* 2

<sup>2</sup>*,* it is a better model and it explains 70% of the total variation of

increases with respect to Model 1

8].

Informal Institutions and Leadership Behavior: A Cross-Country Analysis

http://dx.doi.org/10.5772/intechopen.75840


51









leadership, especially in times of crisis or ambiguity [

moderated by the development of countries. In Model 6,

significant when are moderated with level of development.

*R*

indicating that in terms of

**5. Discussion and conclusions**

leadership under an institutional approach.

\*\*\*

p < 0.01.

leadership behavior is more sensitive to informal institutions such as tolerance. In other words, this negative interaction term indicates that the relationship between tolerance and leadership is stronger when there is a lower, rather than higher, level of development. This is in line with leadership literature that emphasizes the importance of tolerance and stress management in leadership, especially in times of crisis or ambiguity [8].

Although not fully synonymous, the intolerance of ambiguity (an individual cognitive state) and uncertainty avoidance (a behavioral phenomenon) are concepts that are likely to be positively related [92]. In Model 3, we presented a model with the interaction terms between social capital and development. In this model, the coefficients are positive and statistically significant for tolerance, creativity and responsibility. Otherwise, they are negative and statistically significant for power and not significant for social capital. The interaction of social capital and development with leadership was not significant. In Model 4, tolerance, social capital and responsibility were positive and significant and power was negative and significant. Contrary to expectations, the direct effect and interaction term for creativity and the level of development is not significant. Similar results have been found in studies such as Dubinsky et al. [93]. They argue that contradictory results could be explained because creativity intelligence seems vague or unformulated. Similarly, Model 5 presented the interaction term between power and development. Although the main effect of power was negative and significant, the interaction of power and development on leadership was not significant. Finally, in Model 6, we can see the interaction of responsibility and development with leadership. Although the main effect of responsibility was negative and significant, the interaction effect of power was positive and significant. This interaction means that countries with high levels of development experienced a stronger positive impact of responsibility on leadership. As we can see, all models confirm the importance of informal institutions, especially when these institutions are moderated by the development of countries. In Model 6, *R*<sup>2</sup> increases with respect to Model 1 indicating that in terms of *R*<sup>2</sup> *,* it is a better model and it explains 70% of the total variation of leadership. Our results do not support Hypothesis 6, since not all informal institutions were significant when are moderated with level of development.
