But he and his fellow researchers are confident, he said, that the number of companies that backdated at least some options is close to the total reached in their study.
He also noted that this number -- around 1,400, or 43 percent of the publicly traded companies that granted options from 1996 to 2002 -- was in line with estimates reached by other academic researchers.
In an interview, Professor Bizjak said that though this hurdle was quite high, surpassing it wasn't proof that a company backdated options.
There is a small theoretical possibility, he said, that the company was just lucky in picking very favorable days for granting at least 20 percent of its options.
According to the study, a big part of the answer may be found in interlocking boards, which the researchers describe as the biggest conduit for transferring information about options backdating.
Graph tracks percent of companies that may have backdated options* from 1996-2002.
*A company was deemed to be a likely options backdater if the effective dates of more than 20% of the options it granted occurred immediately after a big drop in the company's stock and immediately before a big run-up.
We examine the role of board connections in explaining how the controversial practice of backdating employee stock options spread to a large number of firms across a wide range of industries.
That's because the odds are extremely small, they say, that a company would randomly pick a date so conveniently sandwiched between a large price drop and a subsequent price hike.
Even so, the researchers didn't assume that a company had engaged in backdating unless at least 20 percent of its option grants were suspicious.