Corruption and its diverse effect on credit risk: global evidence

Future Business Journal

Table 9 Robustness test 2 (change of control variables)

We have performed panel least square regression based on the model: \({\text{NPL}}_{it} = \alpha_{i} + \beta_{1} {\text{Corruption}}_{it} + \mathop \sum \limits_{i = 1}^{i} \beta_{2} {\text{Controls}}_{it} + \varepsilon_{it}\). In model 1, the measure of corruption is the CPI score obtained from Transparency International. In model 2, we used the control of corruption as an explanatory variable. In both models, three control variable is retained from the original model which include GDP growth, unemployment and inflation. In addition to these three control variables, five control variables are introduced from World Governance Indicators (WGI) published by the World Bank. Hausman test score indicates that the fixed effect model is appropriate for the study. Therefore, we report fixed effect regression scores in Table 9.
Asterisk ** and *** represent significance level at 5 and 1% respectively.