Direct Democracy and the Effects of Voter Ignorance on Tax Policy Ballot Propositions
IN THIS ARTICLE
Many scholars believe that the mass public is ignorant regarding political matters (Somin 1998; Converse 1964; Campbell, Converse, Miller, Stokes 1980; Caplan 2007). Somin (1998) divides this ignorance into three types. One is a lack of information regarding policy issues, governmental structure, and how the government operates. For instance, majorities in the mass public do not know who their state’s senators are or what branch of government has the power to declare war (Somin 1998, 417). The second type of ignorance is a lack of “an ‘ideological’ view of politics capable of integrating multiple issues into a single analytical framework” (Somin 1998, 417). This results in ideological inconsistency among issues (Somin 1998, 417). Without an ideological view, a person may be pro-choice (regarding abortion) and also support the death penalty. With an ideological view, be it liberal or conservative, a person would know that he cannot support both. The third type of ignorance is the inability of voters “to spot interconnections among issues” (Somin 1998, 418). As a result, a voter may support both cutting taxes and increasing public services without being aware that tax dollars pay for public services, so fewer tax dollars means fewer services. Indeed, as this paper will argue, direct democracy initiatives in the form of tax policy ballot propositions reveal that voters suffer from this third type of ignorance: the inability to spot interconnections among issues.
Section I: Background on Direct Democracy
The idea behind direct democracy is that it is a means for the people to voice their policy preferences. Indeed, direct democracy was largely institutionalized in the early twentieth century by minority groups frustrated by their lack of political representation (Bowler and Donovan 1998, 2). Distrust of political representatives also underlay direct democracy. Oberholtzer (1900) describes representatives as “selfish and dishonest, who would use the government to enrich themselves personally and the class which they represent, the “Boss” and his men who are the curse of the system in America” (392-3). Representatives did not accurately convey the public’s will. Direct democracy initiatives were believed to militate against this problem. In Direct Legislation by the People, populist Nathan Cree (1892) explains, “If it is desired to find the prevailing sentiment of society in order to shape it into law, it is of primary importance to obtain the best mode for it to form and announce itself. The initiative and referendum are the best devices yet suggested to accomplish this end. Under a correct application of them, public opinion can be ascertained and safely accepted as the voice of the state” (119).In other words, direct democracy initiatives are supposed to reflect the masses’ policy preferences (Bowler and Donovan 1998, 4). If this is the case, direct democracy initiatives should also reflect the masses’ lack of political knowledge (as ascertained through their preferences). Indeed, this is one of the reasons why I chose to reveal voter ignorance via direct democracy.
I also chose to use direct democracy because direct democracy initiatives enable citizens to vote on particular issues rather than candidates. Candidates represent bundles of issues: one voter may vote for the candidate based on Issue A, another voter based on Issue B, and so forth (Clark 1998, 450). Put simply, voters may all vote for the same candidate but based on different issues. With direct democracy, the issue on which voters vote is very clear.
There are three forms of direct democracy: referendum, recall, and the citizen’s initiative, commonly referred to as the ballot proposition. This paper will focus on the ballot proposition, which is when groups outside of the legislative arena draft a law and then petition to get the law on the ballot for statewide elections (Bowler and Donovan 1998, 3). A common subject of ballot propositions is tax measures (Bowler and Donovan 1998, 7). I will be dealing specifically with tax related ballot propositions in this paper.
Section II: Theoretical Argument
If it is the case that direct democracy initiatives, and therefore ballot propositions, truly reflect the masses’ policy preferences for some issue, then ballot propositions also reflect an implicit voter ignorance. The type of voter ignorance I am specifically referring to is the inability of voters to make logical connections between idea elements.
Philip Converse (1964) discusses this phenomenon in “The Nature of Belief Systems in Mass Publics.” He states that voters may be unaware that they hold logically contradictory beliefs. Indeed, Converse even cites tax policy as a subject that is particularly prone to logical inconsistencies on behalf of the mass public. Converse (1964) states, “[A]mong adult American citizens, those who favor the expansion of government welfare services tend to be those who are more insistent upon reducing taxes” (209). Somin reiterates this point. An informed electorate, he states, “would not be oblivious to the contradiction between seeking a reduction in government power [i.e. taxes] and an expansion of nearly all its major programs” (Somin 1998, 418).
Converse explains that the mass public is more susceptible to believing logically contradictory things because “simple ‘thinking about’ a domain of idea-elements serves both to weld a broader range of such elements into a functioning belief system and to eliminate logical inconsistencies” (Converse 1964, 209). Since political elites, and educated people in general, think about such things more than the mass public (comprised of ill-informed people), one can expect that the mass public will be more prone to logical inconsistency (Converse 1964, 210). Converse offers an illustration of this phenomenon. He states, “If a legislator is noted for his insistence upon budget balancing and tax-cutting, we can predict…that he will also tend to oppose expansion of government welfare activities. If, however, a voter [a member of the mass public]…casts a vote for him directly out of enthusiasm for his tax-cutting policies, we cannot predict that the voter is opposed as well to expansion of government welfare services” (Converse 1964, 210).
It should be noted that these logical inconsistencies occur independently of party affiliation (Converse 1964, 209). That is, specific party ideologies do not make its adherents logically inconsistent on the issue of taxes and government services.
In sum, tax ballot propositions reflect the mass public’s policy preferences on the issue of taxation. These particular policy preferences are generally inconsistent regarding tax cuts and government services. The mass public wants tax cuts and will vote for tax cuts. Yet their vote for tax cuts does not imply a vote for fewer services. Given their propensity for inconsistency, a vote for tax cuts implies a vote for more services.
Section III: Proposition 13 and Measure Five
California’s Proposition 13 and Oregon’s Measure Five exemplify this theory that a vote on taxes implies a logical inconsistency regarding services. When one studies these propositions and the public opinion data taken after their passages, the mass public’s logical inconsistencies become clear. I specifically chose these two cases because public opinion data regarding governmental services was available.
California’s Proposition 13 passed in 1978. Proposition 13 stated that the property tax cannot exceed 1% of the full cash value of the property (being taxed). 4,280,689 votes were cast in favor of Proposition 13, and 2,326,167 votes were cast in opposition to it (Sears and Citrin 1982, 22). In the year after the proposition passed, property tax revenue fell by more than five billion dollars. California had a surplus of about four billion dollars and this helped mitigate the loss in property tax revenue (Perry and Sonstelie 1982, 120). However, by 1981 the state had a one billion dollar deficit (Perry and Sonstelie 1982, 125).
Data from public opinion polls reveals that Californians wanted an increase in governmental spending on public services before, during, and after the campaign (and subsequent passage) of Proposition 13 (Sears and Citrin 1982, 48).
Figure 1: Preferences for increases or decreases in government spending on specific services.
The May 1978 column consists of public opinion data taken on the eve of voting for Proposition 13. The numbers here are lower than the rest of the columns, indicating that the public’s support for increasing expenditures decreased. It is not clear why this is so. However, according to Sears and Citrin (1982), “[e]ven here those who wanted to maintain the status quo far outnumbered either expanders or cutters, who approximately offset each other” (49). This nonetheless implies a logical inconsistency: if one cuts a source of funding to these services, it does not seem reasonable to expect the level of expenditure for these services to remain the same.
Sears and Citrin also note that those polled supported increased expenditures for universal services. These are services that are intended to benefit everyone, such as safety (police and fire departments, as well as prisons and corrections facilities), education (public schools), and public transportation (Sears and Citrin 1982, 49). Services intended to benefit a particular group of people, such as public housing and especially welfare, generally saw declining levels of support. Interestingly, “universal access to [the aforementioned services] makes them expensive, so we are left with the irony that the public supports expansion of the costliest governmental responsibilities, while simultaneously demanding reduced taxes” (Sears and Citrin 1982, 49).
Oregon’s Measure Five passed in 1990. Measure Five set a 1.5% property tax rate ceiling that would be achieved after a five-year phase in period. Unlike Proposition 13, which passed with a wide margin of support, Measure Five passed with a much narrower margin of support. 574,833 votes favored Measure Five while 522,022 were opposed (O’Toole and Stipak 1998, 9). In the two years following the passage of Measure Five, property tax revenue fell almost five percent. The property taxes that cities, counties, and school districts lost totaled over an estimated $600 million (O’Toole and Stipak 1998, 9).
During this time, researchers O’Toole and Stipak sent out a mail questionnaire to 274 cities and counties in Oregon. The questionnaire was sent to city and county managers, recorders, finance and budget directors, and other officials (O’Toole and Stipak 1998, 10). One of the questions dealt with the perceived level of citizen expectations regarding the provision of public services. According to the data, there was a 6% decrease in citizens’ expectations for more state services while there was a 56% increase in such expectations. Actual state services levels increased by 24% and decreased by 40% (O’Toole and Stipak 1998, 10).Continued on Next Page »