The striking thing about the Carter Center's "reply" to Hausmann and Rigobon's original and much ballyhooed claims is that, well, it's not really a reply at all - it's just a reassertion of what they'd said before. The CC's report made no effort to engage with the substance of Hausmann and Rigobon's argument, leaving the back-and-forth to look very much like a dialogue of the deaf.
This is particularly vexing given that Carter Center could have leaned on a massive, gaping hole in Hausmann and Rigobon's argument - call it Dan's Really Obvious Objection. Kudos have to go to Dan Burnett who spotted this almost two weeks ago and wrote about it in the discussion board here: if, as Hausmann and Rigobon argue, the Cold Audit was carried out on a random sample of a random sample of untampered with voting centers, how can its overall results possibly match CNE's supposedly fraudulent results?
Now CEPR - a propagandistic philochavista "think tank" in Washington DC - picks up this argument and uses it as a bludgeon to hammer at the Hausmann and Rigobon paper. Personally, I can't hide my extreme distate for a pseudo-independent outfit like CEPR - which hides an extremist ideological agenda and a clearly partisan stance behind the guise of a properly sanitized DC research center.
But one thing I can tell you: the sky doesn't stop being blue just because an extremist nut says the sky is blue.
In this particular case, Hausmann and Rigobon appear to have pitched CEPR such a juicy bombita that it's hardly surprising they've hit it hard. If the cold audit was carried out on a random sample of "clean" voting centers, CEPR calculates, the chances of the audit yielding results in line with CNE's overall results if the real result was in line with Sumate's exit poll comes out to 1 in 28 billion trillion.
This is just a mathematically retelling of Dan's Really Obvious Objection. What vexes me most is that I know that Rigobon personally had this argument put to him - and rather than giving any kind of reasoned response, flailed his arms a lot saying one could not extrapolate from the audit sample to the entire population. But they knew, they had to know, that Dan's Really Obvious Objection was coming. They don't seem to have had a response for it. If they did, they should've put it in the public domain long ago, before CEPR made the point. If they didn't, then they shouldn't have published their paper at all.
Join a moderated debate on this post.