Where the good signatures are: Signature collection and initiative qualification in California

We wish to thank Alex Chang for his assistance with data collection and Shaun Bowler for his comments on earlier versions of this research. We also thank John Mott-Smith and the California Secretary of State's Office for providing us access to the data we use in this paper.

Abstract

Using data from eight recent California initiatives and data provided by the California Secretary of State's Office, this research explores the geographic source of signatures and their distribution across counties, investigates both total signatures and valid signatures, and presents a regression analysis to study how characteristics of counties relate to the number of signatures gathered. The findings indicate a high rate of equity in the distribution of signature gathering and little demographic targeting across measures. The study also finds that a variety of social and political factors influence the number of total and valid signatures across counties, though the results are more consistent for the former.

Keywords:

• Direct democracy
• Initiatives
• Signature collection

Notes

We wish to thank Alex Chang for his assistance with data collection and Shaun Bowler for his comments on earlier versions of this research. We also thank John Mott-Smith and the California Secretary of State's Office for providing us access to the data we use in this paper.

1 Arguments on one side suggest that the initiative process is dominated by wealthy business interests (Ellis 2002; Smith, 1998); others observe that money is rarely enough to successfully pass an initiative and that the benefits of the process still accrue mainly to broader-based citizen groups (Boehmke, 2005a; Ernst, 2001; Gerber, 1999; Matsusaka, 2008).

2 For more information, see http://www.sos.ca.gov/elections/ballot-measures/history-initiatives-info.htm.

3 An excellent extended discussion of signature gathering and the issues surrounding it is contained in Ellis (2002, Chapter 3). Briefer discussions are contained in Broder (2000), Cronin et al. (1989), Ellis (2003), Lowenstein & Stern (1989) and Magleby (1984, 1985).

4 Also, other research shows that how and what people read and write affects their perception of its content and quality (Dukes & Albanesi, 2013).

5 Oregon first attempted to ban them in 1909; WA, SD, and OH banned them about five years later. These bans have generally been overturned by the courts.

6 Personal communication, Shaun Bowler, 2004.

7 One circulator admitted that people were more likely to sign if he told them it was his birthday (Ellis, 2003).

8 For a well-known example, see Ed Koupal's description of gathering signatures using the “hoopla process” in Cronin et al. (1989).

9 Other forms of direct legislation in California include the popular referendum and recall, both of which require a certain number of signatures. While the process is almost certainly the same, this paper focuses on the far more common statutory and constitutional measures.

10 Originally, this requirement was 8%, but was dropped in 1966 due to flagging usage.

11 A typical distribution requirement exists in Nebraska, where a 5% signature requirement must be met in at least two-fifths of its counties.

12 In cases where there are less than five hundred total signatures submitted, all are checked.

13 For the relevant period of 2000 and 2003, data is missing data for two of the six constitutional amendments that made it to the ballot (Propositions 38 and 39, both in the 2000 general election) and one of the five statutory proposals (Proposition 50 in the 2002 general election). Proposition 38 focused on school vouchers and received only 29.4% of the vote, while Proposition 39 regarded lowering the voting threshold for school bonds to 55% and received 53.4% of the vote. Proposition 50 concerned water quality, water projects, and wetland protection and passed with 55.3% voting yes.

14 In the course of communications regarding other research, officials in the Elections Division indicated that they retained some data on signature checks for some recent ballot measures. Upon request, they provided a spreadsheet with data on the eight ballot measures, and only these eight.

15 The table reports the results of the full checks for these measures rather than the sample checks. A random sample validation result for one petition is unavailable.

16 Even if it did, no states have distribution requirements that require at least one signature from every county (or district in some cases).

17 Note that the estimated validity rates compare favorably to the rates obtained by the Ohio Secretary of State's Office in a 1983 study. The petition circulated by professionals had a validity rate of 68%, and the two petitions circulated by volunteers had a validity rate of 84% (Lowenstein & Stern, 1989).

18 These data are available for public viewing at http://cal-access.ss.ca.gov/Campaign/Measures/.

19 Excluding Los Angeles, the correlation is 0.92 and the slope coefficient is 1.10.

20 The Gini index for the number of signatures per capita in each county varies from 0.38 to 0.50.

21 California Commission on Campaign Financing, quoted in Broder (2000).

22 While there are not many analyses of initiative behavior at the county level, these control variables substantially mirror those used by McFerrin and Adkisson (2012) in their study of county level voting patterns across California initiatives.

23 Note that only the density variable is not robust to excluding Los Angeles County from the analysis.

24 Using a statutory initiative, Proposition 830, as the baseline case, the magnitude of the first differences is greater for constitutional proposals not because the variables matter more, but because more signatures must be gathered for the latter.

25 These models were also estimated using grouped logit, which is appropriate when the dependent variable is the proportion of successes among all trials in acounty. The results were similar in terms of sign and significance.

26 An additional variable that was not included is the proportion of signatures gathered by volunteers versus professionals. Unlike California, Oregon requires circulators to report this figure. The Ohio Secretary of State's study and statements by professional circulators suggest that volunteers have a higher validity rate (Ellis, 2003; Lowenstein & Stern, 1989).