This study aims to find out, analyze and measure the direct and indirect influence of leadership and competence on the performance of local governments with innovation as a mediation variable. The total sample in this study was 486 respondents. The data were tested using SEM PLS by combining models
formatively and reflectively. Performance Measurement in this study is different from what has been done before because it uses measurement indicators formatively. The results of the seven hypotheses proposed obtained the result that five hypotheses were accepted and two hypotheses were declared rejected. The hypotheses rejected are the direct influence of leadership and on innovation, as well as the indirect influence of leadership on the performance of local governments through innovation. This research has supported previous research that although the method of measuring the performance of local governments is different from before, it is still found that the performance of local governments is still based on outcomes. The results showed that the variable that has the greatest direct influence is innovation but innovation is not able to mediate the indirect influence of leadership on the performance of local governments. However, innovation is able to mediate the influence of competence on the performance of local governments. The
results of the study refute the influence of leadership on innovation, although it is said that innovation depends on the ruler but this cannot be proven because innovation depends on the budget and rigidity of the local government bureaucracy.