Josh and I thought we would collaborate for a post on one of this election’s direct democracy initiatives in California, Proposition 19. He and I, along with Dan Smith, have a paper coming out in American Politics Research on the link between statewide ballot measures and congressional roll-call behavior. We knew based on prior work that the federal government often responds to the passage of statewide ballot measures, so we thought it plausible that members of Congress might alter their voting on legislation that parallels ballot measures in their state. Theoretically, we were interested in whether the precise information revealed by a ballot measure—in particular the inimitable insight into their state or district’s median voter policy preference—helps reduce policy shirking—“legislative behavior that differs from what would be observed given perfect monitoring and effective punishment by constituents (Rothenberg and Sanders 2000: 316). In other words, we think that “As information-seekers, there is good reason to think that some lawmakers will value the precise information that ballot measures provide about their constituencies, which is more detailed than that gleaned from polling, prior election results, or a lawmakers ‘general intuition.’” Ultimately we thought this was an interesting question pertaining to federal representation (for an excellent overview see Uslaner’s The Movers and the Shirkers). After examining three cases (campaign finance, gay marriage, and raising the minimum wage) our results show that ballot measures are indeed an important component of roll-call voting in the House. Depending on model specification, representatives were between 56% and 83% more likely to vote in with their district’s median-voter proceeding a ballot-measure. We find this effect is statistically meaningful even when including a number of strong controls (for example their vote on a similar proposal at an earlier time period). Interestingly, we find that the passage rate of the ballot measure fails to influence representatives voting behavior (i.e. what percentage of voters voted to approve/reject the measure). We argue in the paper that this indicates information beyond the ideal point of a member’s median constituent is actually background noise and confounds the effect (but we also admit a measurement issue pertaining to the inherent nature of binary choice data). In the Senate, however, we find no evidence for a link between roll-call behavior and ballot measures. This is not entirely surprising; research has shown that Senators are less reactive to public opinion compared to members of the House.
With an election looming questions about how statewide ballot measures might affect politics (broadly) have been posed. Nowhere is this more salient than in California; where Proposition 19, if passed, will legalize marijuana. Polling over the year has found public opinion largely split on the measure; though most (if I recall) show a narrow plurality in favor of Prop 19. Of course a recent Newsweek poll shows a narrow plurality opposed to the proposition (standard caveats about margin of error apply).
Should marijuana legalization become a legitimate agenda item in Congress, Proposition 19 could be a cue for members of California’s House delegation. While there haven’t been any formal votes in Congress on the issue, Barney Frank (D-MA) and three cosponsors (including Dana Rohrabacher (R-CA)) introduced a marijuana legalization bill in 2009 (the “Personal Use of Marijuana by Responsible Adults Act”). Their proposal would legalize marijuana and tax it at a particular rate. Frank also introduced a medicinal marijuana bill (HR 2835) allowing doctors nation wide to prescribe marijuana for medical purposes. Currently both bills are under consideration, which can easily be translated into, “will never see the light of day.” We’re not going to get into the merits of these bills. Regardless, couple this with the fact that 13 states have decriminalized marijuana, and the thought of these bills reaching the floor in the near future does not seem so far fetched.
Thus, it is possible that Proposition 19 will provide a “moment of clarity” for these members and reveal in fairly unambiguous terms what their constituents want from them. A recent poll conducted by Newsweek suggested that there is a surprisingly high number of Americans (45%) who claim to support California’s legalization efforts. Maybe we’re out of the loop on national public opinion on this issue, but this seems high to us. Moreover, we would suggest that this figure actually underestimates the true percentage. Respondents admitting they favor marijuana legalization may lie to avoid the interviewer’s judgment (aka social desirability bias). Here is the actual question wording and results:
Now moving to a different topic.... As you may know, California's Proposition 19 would legalize marijuana in the state, allowing people age 21 or older to possess and cultivate small amounts for personal use. It would also allow local governments to regulate and tax commercial production, distribution, and sales of marijuana. Would you support or oppose a similar measure to legalize marijuana IN YOUR STATE?
With a narrow split nationwide (also see question #11 for the results on nation wide legalization), this begs the question: How many members will change their prospective vote on marijuana legalization should it come up in Congress? Of course, we can only speak to the California delegation. But considering the current poll, along with the prospective vote share for the ballot measure in each district, we roughly estimate that 3-4 legislators previously opposed to marijuana legalization would adjust their positions on the issue. And this is in California alone. Obviously, states without direct democracy (initiatives and popular referenda) were omitted from our study and likely immune to this kind of pressure. Since other forms of direct democracy require that the state legislature propose the bill or amendment to the people, it lacks the popular impetus to consider the bill.
Our point in this paper, and in this blog post, is that direct democracy results can make their way into congressional voting decisions. Whether this is enough to pass a bill in Congress is anyone’s guess. The social forces preventing legalization are far stronger than America’s other “prohibitive movements.” America only lasted 13 years in Prohibition before the “wets” outnumbered the “drys” and Congress repealed the 18th Amendment. Popular efforts to keep marijuana illegal have proved far more enduring.