Of the scholarly journals started from 2000 - 2006 recorded in Ulrich's, the open access journals were ten times more likely to be still active, strongly suggesting an open access survival advantage for new journals.
Data from Ulrich's, July 5, 2007:
# of online, refeered, scholarly / academic journals started 2000 -
# of above ceased: 59 = .026%
# of online, refereed, scholarly / academic journals, open access
journals started 2000 - 2006: 724
# of above ceased: 2 = .0027
The period 2000 - 2006 was selected, to help control for older, subscription-only journals that would have ceased before open access was an option the journal would have considered.
It should be noted that this is a quick study, which has not explored or controlled for all variables; conclusions should be drawn with caution. The data do, however, strongly suggest an hypothesis worthy of testing.
This post emerged from a discussion on the potential positive spiral in transition to open access on Liblicense. It is part of the Transitioning to Open Access series.