The mandatory constitutional convention question referendum: the New York experience in national context.

• Author: Benjamin, Gerald

THE MANDATORY CONSTITUTIONAL CONVENTION QUESTION REFERENDUM: "NO" IN NEW YORK IN 1997

Voters expressed little confidence in government; turnout at the polls was consistently abysmal. Legislative elections rarely offered real choices; incumbents almost never lost. Gridlock was the norm in a state legislature that featured the most persistent divided partisan control in the nation. The state budget had not been passed on time in thirteen years. (1) The state personnel system was sclerotic. A torturous local government web--a "system" in name only--diffused accountability and drove up costs. State and local taxes, especially local property taxes, were among the highest in the nation. (2)

The result of all this was a state and local service delivery system that was expensive, inequitable, and often inadequate. Education is the best example. Mean per pupil education spending was very high. (3) Children in the suburbs were well served, or at least had a fighting chance. But most children--especially minority children in urban centers--were simply not being educated. (4)

Yet, when asked in 1997, in the midst of these conditions, to vote on the question "Shall there be a convention to revise the constitution and amend the same?" New Yorkers responded with a resounding "No." The vote was 929,415 in favor of a convention, to 1,579,390 against. (5) Perhaps even more tellingly, a plurality of citizens who came to the polls in that year--1,693,788 of them--simply ignored the question entirely! (6) The idea of holding a convention was rejected even though Governor Mario M. Cuomo had earlier endorsed it as the state's best chance for reform; (7) even though the commission he appointed worked for several years to prepare for it; (8) and even though by the time of the vote virtually every daily newspaper in the state had published an editorial in favor of holding a constitutional convention. (9)

The convention question was on the ballot in 1997 because a century-and-a-half earlier (in 1846) a Convention in New York added a constitutional requirement that the question of whether to call a convention be asked every twenty years. (10) The idea for a mandatory convention referendum at regular intervals first appeared in the late eighteenth century in the constitutions of Massachusetts, New Hampshire, and Kentucky. (11) The Empire State is currently one of fourteen in the United States whose constitutions require the periodic submission of such a question. (12) Perhaps because the idea was included in the Model State Constitution, (13) many of these states adopted the provision relatively recently: Alaska (1956), Connecticut (1965), Hawaii (1950), Illinois (1970), Michigan (1963), Missouri (1945), and Montana (1972). (14) Additionally, Rhode Island added the periodic convention-call provision to its constitution in 1973. (15)

One rationale for such provisions is that the sovereign people should have some way of making changes in their governmental structure without having to rely on action by those in statewide and legislative offices, many of whom may be beneficiaries of a flawed status quo. Another is the Jeffersonian view that it is healthy for democracy for each generation to define anew its governing arrangements. Thomas Jefferson wrote in 1816 that "`Each generation [has] ... a right to choose for itself the form of government it believes most promotive of its own happiness.... [Al solemn opportunity of doing this every nineteen or twenty years should be provided by the Constitution.'" (16) A third, more conservative reason for these provisions is that periodic convention votes are a way of actually testing public support for political reform ideas, and of simultaneously channeling political energy and "avoid[ing] agitation." (17) Such referenda are more likely to confirm the status quo than to result in conventions actually being called, this view holds.

New York's failure to authorize a constitutional convention through an automatic convention question referendum is hardly unusual. In a comprehensive review published in 1970, Robert J. Martineau found that there were seventy-two votes resulting from the automatic convention referendum provisions of state constitutions between the founding of the nation and 1969. (18) Twenty of these (27.8%) led to the calling of conventions in five states: Michigan, Missouri, New Hampshire, New York, and Ohio. (19) Yet more than half of these conventions--eleven of the twenty--were in New Hampshire, which until recently provided for no means other than a convention for amending the state constitution. (20)

Since 1970 there have been twenty-five additional referenda resulting from automatic call provisions. (21) Four produced conventions: two in New Hampshire and one each in Hawaii and Rhode Island. (22) Thus, the success rate during the past three decades (16%) has been substantially lower than in the past. (23) No conventions have been authorized by voters under an automatic call provision in the fourteen states since 1984. (24) A 1996 referendum in Hawaii produced a supportive majority of those voting on the question, but no constitutional convention was held. (25) Litigants who claimed that the required majority had to be of all those who voted at that election were supported in the courts. (26)

Despite the mandate in its state constitution that the question be asked every twenty years, no vote was held in Oklahoma in 1990 on whether to call a convention. (27) Janice C. May reported that "the legislation necessary to place the referendum on the ballot did not clear the legislative process and no vote was taken." (28) Failure to provide for balloting on the question is only one of the ways state legislatures have sought to block conventions by inaction. Another is by failure to authorize the preparatory work to educate the public on the importance and meaning of the convention vote, and then arguing that in the absence of preparation a convention would be too risky. (29) A third is by failure to provide for the election of delegates or for the logistical support necessary to hold a convention.

Because the New York State Constitution prescribes the precise question to be asked the voters--"Shall there be a convention to revise the constitution or amend the same?" (30)--the agenda of a constitutional convention in New York may not be limited. The situation is similar in eight other mandatory question states. (31) Six mandatory convention states also allow constitutional amendment through an initiative--a more targeted method for bypassing those in power to make change. (32) The availability of such an option may make the convention route to constitutional change even less attractive.

The inability to limit a convention's agenda makes gaining the cooperation of the state legislature--termed "indispensable" by a team of political scientists who comprehensively reviewed the extensive efforts at state constitutional change in the 1960s-extremely problematic. (33) Legislatures began as the dominant governmental institutions in the separation-of-powers systems of the American states. (34) Constitutional change over more than two centuries has, in general, been a stow of the diminution of the role and powers of legislatures. It is no surprise, then, as Albert Sturm noted, that legislatures, as the principal "repositories of general policy making authority," are natural enemies of unlimited constitutional conventions. (35)

TIMING AND ELECTION CYCLES

Predictable electoral cycles and fixed decision points are a defining characteristic of American politics. Public officials run for offices with fixed terms, usually two or four years; election day for most national, state, and even local offices is the first Tuesday after the first Monday in November. Longer cycles are defined by term limits--traditionally for statewide elected officials--but, lately in many states, for legislators as well.

The convergence or divergence of these cycles affects voter turnout, ballot length, the availability of campaign resources, and a number of other factors that may impact political outcomes in a particular election year. Often, structural arrangements are made to limit or eliminate specific convergences. In thirty-five states with four-year terms for governor, gubernatorial elections are held in even-numbered years that are not presidential election years. (36) In Virginia, New Jersey, Kentucky, Louisiana, and Mississippi gubernatorial elections are held in odd-numbered years. (37) To cite a less-known example of structural arrangements made to limit or eliminate specific convergences, the New York State Constitution provides that city-elected officials be chosen in odd-numbered years, and that all their terms expire in odd-numbered years. (38) Peter Galie writes that this was intended by Progressives in 1894 as a "home rule" provision that "separated state and national elections from municipal elections so only municipal issues would determine the outcome." (39) Of course, it also sheltered elections for state offices from the turnout that might be stimulated in New York City and other big cities by convergent mayoral elections.

The mandatory constitutional question provision adds another long cycle to fourteen states' political systems. There are clear effects when this cycle converges with the cycle for other elections. For example, a comparison of the numbers voting in Iowa, Alaska, and New Hampshire in presidential and non-presidential even-numbered years--three states whose cycles result in periodic referenda in both--confirms the common-sense expectation that convergence with relatively high-turnout elections result in higher numbers of citizens voting on the mandatory constitution question. (40)

As the prime players, incumbents in state elective offices are keenly aware of the potential that a constitutional convention could change the fundamentals of state politics and government. These...