The Atlantic's McArdle Problem.

A colleague points me to this, from a WaPo chat with The Atlantic’s Megan McArdle:

Anonymous: You said that medical innovation will be wiped out if we have a type of national health care, because European drug companies get 80% of their revenue from Americans. Where did you get this statistic?

Megan McArdle: It wasn't a statistic--it was a hypothetical.

However, whenever I have been able to find pharma financial statements that break down their profits by region, the lion's share always comes from the US.

Before we get to why this is worth taking note of, let’s be clear that on the factual question, she’s simply wrong. As my colleague noted:

As a statement about geographical breakdowns of revenues, and assuming "the lion's share" means "at least half", this is not true about Pfizer, GlaxoSmithkline (pharmaceuticals sales), Sanofi-Aventis, Bayer (healthcare division) or Hoffman-LaRoche. It is true about the "Pharmaceuticals" division of Johnson & Johnson, but not about the "Medical Devices and Diagnostics" division (combined, the US is almost exactly half of sales, 53%).

As a statement about operating profits (the original question was about revenues but the answer talks about profits), it is not true about GlaxoSmithkline (on a location of subsidiary basis rather than final customer basis) or Sanofi-Aventis; the rest of them don't give the operating profit split, but the margins ex-USA are not noticeably lower.

It is not difficult to look these things up; all the big 6 pharma companies have their report and accounts on the web, and all of them have searchable pdfs in which one can find the geographical breakdown of sales really quickly.

Here’s the thing. If I were the editor of one of the most important magazines in the history of this fine nation, which has been publishing since 1857 and probably has to rent out a separate facility to house all the awards it has won, and I was thinking of hiring an economics columnist, I might see if I could find someone who understood the meaning of terms like “statistic” and “hypothetical.”

More after the jump.

--Paul Waldman

You need to be logged in to comment.
(If there's one thing we know about comment trolls, it's that they're lazy)