'Richard Tol finds many points to question. The data is not what was reported, it fails validity tests, does not represent the literature it surveys, saying: “the main finding of the paper is incorrect, invalid and unrepresentative”. Cook’s paper essentially relies on only 12 reviewers, who were not tested for rater bias or rater fatigue (despite answering up to 4,000 questions). There is no survey protocol. There were changes to the test that are not documented. Fifty seven percent of the data remains unreleased. Cook claims “confidentiality”. Tol repeatedly offers to sign a confidentiality clause.'