That's the interesting line that came up in a news article this afternoon. It was just dropped there, as a given. I wondered where this came from, and did a search of the phrase "spend twice as much on marketing." It came up many times.

What I found interesting was that all of it went back, through one source or another, to one paper written by Gagnon and Lexchin.

I find this "inverted pyramid sourcing" quite often. One paper comes out, and in a week, if it says what everyone wants to hear, the conclusion becomes The Conventional Wisdom.

So, I take a look at the methodology. You can read it yourself under "results," but essentially it consists of this: In every case where there are two conflicting numbers, take the larger one, even if the sources switch back and forth on which was larger. Then they tacked a 30 percent markup on it. After all, they knew what results they wanted.

The table included makes this clear:

The $57.5 billion is compared with the $29.6 billion described as research. Note that if you go the other way, with the smaller estimates, the total ends up at $34.8 billion, even with the markup. Basically equal to the research number, and no fun at all. Maybe the real number is somewhere in between.

This is how it's done. Nobody's going to read the paper. Everybody's going to quote the people who quote the sources who quote the paper, and it will be enough.

Won't change anything, but it was interesting.