Have special interests subverted the initiative process?

Over at the Volokh Conspiracy, Gail Heriot challenges the notion that the initiative petition process has been subverted for the benefit of powerful special interests. She cites a soon-to-be-published article analyzing the subversion hypothesis:

It seems that there are real differences in the fiscal policies of initiative and non-initiative states. Initiative states spend less than non-initiative states. Initiative states concentrate more of their spending at the local level. And initiative states raise a greater portion of their revenue through fees rather than through taxes. The subversion hypothesis, however, gets no support from Matsusaka's research. In each case, the initiative states move public policy in a direction that it consistent rather than inconsistent with popular will. Voters tend to want their state governments to spend less money, etc. Hence, instead of subverting the true popular will, the initiative process appears to be giving that popular will a means with which to influence public policy.

I think that is certainly true here in Oklahoma. Initiatives are not often used -- most of the ballot questions we get are referenda from the legislature, required in order to approve constitutional amendments. Only 366 initiative petitions have even been submitted since statehood, and many if not most of those have been ruled legally or numerically insufficient. There appear to have been fewer than 10 over the last 10 years, and it looks like about half never made it to the ballot.

But initiatives gave us term limits, a ban on cockfighting, and set the bar higher for tax hikes -- all issues with popular support, where there was insufficient political will or clout to accomplish them in the legislature. One somewhat recent initiative was clearly the work of a special interest group -- that was the petition to legalize casino gambling, which went before the voters in February 1998. The original sponsors of the drive lost interest once it was on the ballot and it lost by a three to one margin. If, say, a company tried to use an initiative petition to give itself an indirect advantage over the competition, voters would pretty quickly see through the effort. The company would find it a lot more cost effective to lobby 149 legislators than a million voters.