Too Much of a Good Thing? Initiatives and the “Cluttered” Colorado Constitution

State constitutions can affect fiscal policy either by acting as fiscal restraints that limit the scope of government or by imposing fiscal pressures that expand or place demands on government. In 18 states, it is possible to amend the state constitution through an initiative process: voters can petition to put amendments on the ballot directly, bypassing the legislature (which can also place measures on the ballot for voter approval). In states that allow this alternative process, approximately 14 percent of all constitutional amendments have been adopted via the initiative process (initiated amendments) as opposed to being referred to voters by the legislature (referred amendments). States that allow initiated amendments tend to have longer constitutions with more enacted amendments (see table 1). Here we examine the consequences of initiated amendments for state fiscal policy.

Constitutional Amendments and Policy Conflict

The consequences of longer constitutions and a greater propensity to amend the constitution are twofold: First, initiated amendments are another way for interest groups and voters to treat constitutions as statute books, rather than let the constitution create a general structure for governance and leave policy matters to the legislature. As two analysts put it, “The malleability and specificity of state constitutions invite political attention from political actors who otherwise would fight things out in the legislature.” They give the example of California’s Proposition 98, which directs a fixed share of the budget to K–12 education, and thus makes “K–12 a priority in the budget process and reduces teachers’ bargaining uncertainty in negotiating salaries and work rules.” Policymaking by constitution, therefore, limits legislative flexibility, and the availability of the initiative process encourages this approach.

Second, as constitutions increase in length, there is a greater chance that provisions in the constitution will conflict. Direct conflicts occur when two provisions cannot both be satisfied simultaneously. In addition to making policymaking more difficult, conflicts give great power to the judges who are often tasked with resolving them. In Nevada, for instance, a 2003 budget battle between the governor and the state legislature led to a series of court decisions with judges noting the “conflict among several provisions of the Nevada Constitution” and the need for the court to “reconcile the provisions which cause the present crisis” (these provisions included an initiated amendment requiring a supermajority legislative vote for revenue measures and a constitutional mandate that the state fund education). Some provisions, while not in direct conflict, limit policy flexibility by establishing competing objectives. In the discussion that follows, we will give an example from Colorado of one such indirect conflict.

The Constitution and “Haphazard” Fiscal Policy in Colorado

As of January 1, 2016, 50 of the 154 constitutional amendments ever adopted in Colorado were initiated—the third most of any state, whether in terms of the total number of initiated amendments or in terms of initiated amendments as a share of all amendments. Until recently (as discussed later), Colorado also had a relatively low threshold for getting measures on the ballot and required a simple majority for ratification.

Initiated amendments have been on Colorado’s reform agenda for several years. In 2007, a state commission raised concerns about “constitutional clutter” and “conflicting fiscal provisions” owing in part to the constitutional amendments enacted through the initiative process. In 2005, another commission noted, “A most serious issue facing Colorado is its tangled fiscal policy process. Central to the problem is the practice of making fiscal policy by public referendum through amendments to the Colorado Constitution. It is a haphazard approach where citizens are asked to make major fiscal decisions in isolation.”

The authors of the 2005 report also pointed to the “fiscal vise” on state budgets created by the interaction of three constitutional amendments, two of which were initiated and one of which was referred. In 1982, Colorado voters approved the legislatively referred Gallagher Amendment. The Gallagher Amendment placed significant constraints on residential property taxes. In 1992, voters approved the initiated amendment known as TABOR, “The Taxpayer Bill of Rights.” Provisions in TABOR included mandatory referendums on property tax increases and revenue limits on local school districts as well as state governments. Partly because of TABOR, voters in 2000 enacted the initiated amendment known as Amendment 23, which required statewide base per pupil spending to increase by 1 percent plus the rate of inflation through 2011, and increase by the rate of inflation thereafter.

Given the Gallagher Amendment and TABOR, the passage of Amendment 23 effectively prioritized education over other types of spending and placed state legislators in the position of either cutting other government programs or finding a way to increase revenue. In 2005, voters approved a referendum waiving parts of TABOR for five years, thereby giving the state some breathing room budget-wise. In 2010, legislators chose a third approach when faced with budgetary stress: cuts to education elsewhere to counteract the mandated increase in base education funding. In 2015, Colorado’s Supreme Court ruled that this action was constitutional.

The debates over how much to spend on education, whether those costs should be borne at the state or local levels, and how to prioritize education spending vis-à-vis other government programs are important. But these are debates over policy matters and are complicated by the competing provisions in Colorado’s constitution that, on the one hand, constrain tax revenue, and on the other require more spending in certain areas.

In isolation, a limit on tax revenues or the scope of government may be perfectly appropriate. The issue in Colorado’s case, however, is that general fiscal rules clashed with policy-specific spending requirements. As David Primo has written, “Of course, fiscal rules also influence policy, albeit indirectly. A budget rule that limits total spending, for instance, may give an advantage to some policy areas over others. However, although the practical effect of the rule may be to advantage some spending priorities over others, this outcome is not guaranteed if preferences over the distribution of spending change. In contrast, policy-specific provisions in state constitutions allow for less flexibility as preferences or circumstances change, as legislators have learned the hard way with regard to education and pensions.”

Concerns about initiated amendments led Colorado voters to approve Amendment 71 in November 2016. Advocated by the “Raise the Bar” campaign, Amendment 71 changes the thresholds both for placing an amendment on the ballot and for enacting it. To place an amendment on the ballot, petitioners must now secure the signatures of 2 percent of registered voters in each state senate district in addition to meeting the previous requirement (still in effect) of 5 percent of votes cast in the previous election. For an amendment that adds to the constitution, a 55 percent threshold is now required instead of a simple majority. Consistent with the idea that it should be easier to shrink rather than grow the constitution, there is no change to the voting threshold for approving amendments that repeal part of the constitution. Although these actions do not address existing conflicts, they may help to prevent future ones from arising.

What Other States Can Learn from Colorado

Should other states follow suit and consider reforms similar to Colorado’s that, while still allowing initiated amendments, create a higher bar for proposal and passage? Doing so would seem to preserve the benefits of the citizen initiative—that it makes government more effective by serving as a check on legislators—while at the same time limiting the potential for clutter and conflict among amendments.

Let us look again at table 1. On average, states that allow initiated amendments have proposed and approved more amendments than other states, even if we only look at referred (rather than initiated) amendments. Initiative states, therefore, may be distinct in terms of their overall preference for using the constitution for policy purposes. Researchers studying California’s constitution, for instance, write, “[W]e think it is likely that California’s constitution changes as frequently as it does because it has so much in it that invites changing, and because Californians have come to expect everything but the kitchen sink to be in it. Stemming from a long-standing mistrust of representative government in California, the constitution has been used to express the majority will on various policy subjects and constrain representative government in its powers and freedom to make policy.” However, initiated amendments, if too easy to enact, are likely to exacerbate this tendency for policymaking by constitution rather than ameliorate it. For instance, one estimate is that initiatives in California (statutory and constitutional) have locked in about 30 percent of the state budget.

In closing, initiated amendments can, when interacting with other elements of a state’s constitution, create conflicts among provisions, limit legislative flexibility on policy matters, grant judges policymaking power, and exacerbate the tendency of some states to use constitutions as statute books. These concerns need to be weighed against the benefits of the initiative process, which include more responsive government. Colorado’s recent reform to the initiated amendment procedure, which makes it more difficult to propose and approve amendments to the state constitution, attempts to strike a better balance between the goals of constraining the legislature on fiscal policy and preventing the constitution from being used as a policymaking tool.