I could be wrong, but from looking at the authors data, and the writing, it seems as if the author is "massage" his interpretation of the data to fit his hypothesis (which could very well be valid).
The author suggests "You can draw a straight line: The faster they added fees, the more their revenue fell." While it may be true for the worst performing carriers, this trend doesn't hold with the "least bad carriers" (Alaska fell faster than AirTran but added fees later, Southwest fell faster than JetBlue, but doesn't have fees).
While I do see the point the author is trying to make, and it may very well be valid, I'm not prepared to reach the conclusion "The faster airlines add fees for basic services like checked bags, the faster their total revenue declines.", the data he provided just doesn't show that.
edit:typos