This empirical study analyses the extent to which the Productivity Commission (Commission) relies on different types of evidence in formulating recommendations in a sample of reports. It goes deeper than traditional citation analysis; rather than classifying and counting all material cited in each sample report, it includes only citations that influenced the final recommendations. The findings, which run counter to the rhetoric employed in relation to the Commission's work, reveal the extent to which the Commission relies on non-quantitative forms of evidence, including bare assertions, personal experience, and logical and legal arguments, particularly in reports addressing broader questions of social policy. It concludes with a discussion of the significance of these findings, linking them to Graycar's critique of law reform commissions. As such, it provides a more accurate, but still preliminary, basis for understanding the Commission's methods than that otherwise appearing in the discourse surrounding the Commission's work.