Can decarbonization policy results be detected by simplistic analysis of macro-level statistical data?

We study some simple statistical relations between several sets of macro-level data related to decarbonization of the power sector. The data used in the research was obtained from Eurostat's web site. Besides regressions between the variables, we also ran Granger causality tests to explore whether the expected causation relations between them are recognizable from such macro-level data. Although regressions alone proved to be as theoretically expected, we found causality relations to be sometimes counterintuitive, that is, unexpected. The reason to do such a probing analysis was to show whether the macro-level data gathered by official statistical bureaus can be used to present results of policy implementation measures in a convincing way, with clear indication of causality. This may prove to be important because the measures from an energy policy framework usually incur additional costs on citizens. Without being able to see causal relationship (instead merely a coincidental one, which can be seen from simple regressions) between money they (are supposed to) spend, and the policy results, people may develop opposition against the measures, however good they may be.