Upon reflection, I realized what a stupid mistake I had made in interpreting averages and went back to the source report to see if I had misread it. Indeed I had: contrary to what I stated, the 27.7% and 23.3% averages both exclude the US; they are the weighted and unweighted averages, respectively. While I would argue that the unweighted average (23.3%) is more appropriate for policy comparison, I would no longer characterize the ATF statement in question as "highly deceptive". My mistake.