Does the David Victor study condemn most offsets?

Something weird is going on: all over the climate blogs you can read claims that Stanford's David Victor (sometimes together with Michael Wara) has estimated somewhere from one third, one-half, to two-thirds of all carbon offsets under the Kyoto CDM fail "additionality" and do not cause true additional emission reductions. For example, here:

Program Director David Victor said the upcoming study will question the effectiveness of using offsets as a substitute for a safety valve approach to limit the cost of carbon credits. In addition, Victor said it will show that "between a third and two thirds" of emission offsets under the Clean Development Mechanism (CDM) -- set up under the Kyoto treaty to encourage emissions reductions in developing nations -- do not represent actual emission cuts. Victor is developing the analysis with Stanford University's Michael Wara.

Okay - where's the study? Victor and Wara produced this paper a month after the press release above, but I don't see where the third to two-thirds ineffectiveness is in it. Am I missing something, or is this part of the over-blown anti-offset issue?

Anyway, I'll be off soon for a week or so. Have fun, everyone.

UPDATE: Worth noting that even the Wara-Victor study, while very anti-offset, supports the creation of a Climate Fund to reduce emissions in developing countries, and "Perhaps, with time and attention to creating the necessary monitoring system, the Fund could even yield compliance credits for its donors that would be fungible with allowances in cap-and-trade systems." (Page 21.) Sounds like offsets to me.

UPDATE 2: John Mashey points out in the comments that the one-third to two-third quote might refer to offsets for HFC-23, which originally accounted for over two-thirds of offsets and dropped to about one third in 2008. Victor and Wara say payment so exceeded the cost that it created perverse incentive to create and capture HFC-23. I think John's probably right about where this comes from, but I don't think it's a defensible quote based on the paper. All they do is speculate that useless HFC-23 was created and then captured without ever proving that a significant percentage of HFC-23 was useless, let alone all of it was useless. They even admit on the same page (p. 11) that developing countries did vent HFC-23 prior to the CDM, so the capture would satisfy additionality as to those emissions.

A good test of this argument would be whether total HFC-23 production increased after the CDM mechanism was established. Absent a substantial increase, I don't think the quote's justified. My guess, FWIW, is the quote was in an earlier draft but that the authors decided late in the game that they couldn't justify it and then pulled it out.