Although large numbers of studies have examined the Porter hypothesis, one important issue that remains to be addressed is that why past studies give rise to mixed results. By filling this knowledge gap, our research aimed to provide a fuller picture of the Porter hypothesis. By using the method of meta-analysis, including a total of 58 papers across the world, the findings showed that (1) the overall Porter hypothesis was confirmed—that, in general, environmental regulation had a positive effect on green innovation; (2) by decomposing environmental regulatory types and measures of green innovation, it validated the ‘narrow’ version of the Porter hypothesis—that flexible environmental regulation could foster innovation. Especially, we found the command and control regulation had shown its highest consistency and effectiveness in driving green innovation, whereas voluntary regulation had the highest level of flexibility among all regulatory measures; (3) we also found that heterogeneities of the results were attributed to country type and level of analysis. Overall, by integrating studies with large sample sizes and identifying variations among studies, we offered a comprehensive understanding of the mixed results of the Porter hypothesis and generated more precise and generalizable conclusions than the past studies.