Proposals

"Housing" is an issue. "Allow four-story apartments within a half mile of train stations" is a proposal. "Measure H: Shall the city amend its zoning code to..." is a ballot item. These are three different things.

People can care about an issue without agreeing on any proposal. A proposal can exist for years without reaching a ballot. And a ballot item has exact legal wording that may differ from what people discussed earlier.

My Vote Forward asks every proposal to answer a few plain questions. Who has the power to adopt it? What does it cost? Where does the money come from? What else has to happen first? Where is it now in the process? Without those answers, an opinion on it means little.

Defined terms

Issue. A subject or problem area, such as housing, transit, or public safety. Issues are what people prioritize. See Priorities.

Proposal. A specific, worded course of action intended to address an issue. A proposal can be a bill, an ordinance, a budget line, an administrative rule, a plan, or a draft measure.

Ballot item. A contest or question that has legally qualified to appear on an official ballot, in its final official wording.

Adopting authority. The body or electorate with legal power to make a proposal binding: a legislature, council, agency, board, or the voters themselves where law allows.

Cost estimate. A stated estimate of a proposal's fiscal effect, with its source and date.

Funding source. Where the money would come from: existing revenue, a new tax or fee, borrowing, reallocation, or another government.

Dependency. Anything that must happen, or must be true, for the proposal to take effect: another law, a court ruling, a vote of another body, outside funding, or a technical capability.

Status. The proposal's current position in its process, with a date.

Issue, proposal, ballot item

The distinction matters most at the point where people express opinions.

An opinion on an issue ("I care about housing") is really a priority. An opinion on a proposal ("I support this zoning change") is an opinion in the framework's sense. A choice on a ballot item is a vote. Each step adds precision, and each step can change what people think.

California's initiative process shows how much precision a ballot item requires. Before a statewide initiative can circulate, the Attorney General prepares a "circulating title and summary," which becomes the official summary of the measure. That summary must include, in boldface, either an estimate of the measure's effect on state and local revenues or costs, or an opinion on whether a substantial net change in state or local finances would result. The Department of Finance and the Legislative Analyst must jointly prepare that estimate within 50 days ¹. By the time voters see the measure, its wording, summary, and fiscal estimate have been fixed through a legal process. An opinion given earlier, on an informal description, may not match it.

California also limits what a single initiative may contain: under its constitution, no initiative "addressing more than one subject area may be submitted to the voters or have any effect" ¹. The National Conference of State Legislatures notes that states often restrict initiative topics through such single-subject rules ². These rules exist partly so that a voter's single choice can map onto a single question.

Hypothetical example. A neighborhood survey asks residents whether they support "better transit." Eighty people say yes. Later, a proposal appears: a new bus line funded by a quarter-cent sales tax. Some of those eighty oppose the tax. Their earlier "yes" was about the issue, not the proposal. Reporting it as support for the proposal would be wrong.

Authority: who can actually decide

Every proposal must name the body that could adopt it. This is often not obvious.

• A city council may control zoning but not the regional transit agency.
• A school board may set curriculum within state rules but not state funding formulas.
• A state legislature may be blocked by its own constitution, which only voters can amend.
• Voters may be able to adopt some laws directly, but only where law provides it.

Direct lawmaking by voters exists in some places and not others. NCSL reports that 24 states, the District of Columbia, and the U.S. Virgin Islands have an initiative process; 23 states, the District of Columbia, and the U.S. Virgin Islands have the popular referendum; and legislative referenda "may appear on the ballot in all 50 states" ². In Switzerland, a federal popular initiative needs "the signatures of 100,000 voters who support the proposal" collected within 18 months; an optional referendum on a new federal law needs 50,000 valid signatures within 100 days of its publication; and all constitutional amendments approved by Parliament go to a mandatory referendum ³.

The framework does not assume that any proposal can be put to voters. It asks, for each proposal, what route to adoption actually exists in that jurisdiction. See Jurisdictions and Participation between elections.

Cost, funding, and dependencies

A proposal without a cost is incomplete. A cost without a funding source is incomplete in a different way: it hides a tradeoff. See Tradeoffs.

The framework asks each proposal to state:

1. Cost estimate, with who produced it and when. Where an official estimate exists, such as California's joint fiscal estimate for initiatives ¹, it is cited. Where none exists, the proposal says so rather than inventing one.
2. Funding source. New revenue, reallocation (and from what), borrowing, or outside funds.
3. Ongoing versus one-time costs. A building costs money once; staffing it costs money every year.
4. Dependencies. Other approvals, laws, or conditions required. For example, a local measure may depend on a state law allowing the local tax.
5. Uncertainty. A range where the estimate is uncertain, and what would change it.

Effectiveness is a separate question again: whether the proposal would achieve its goal. That is assessed with evidence, not inferred from cost or popularity. See Tradeoffs.

Status changes over a proposal's life

Proposals move. They are drafted, revised, merged, withdrawn, qualified, defeated, adopted, challenged, and implemented. An opinion recorded at one stage may not apply at another.

California's process shows how formal these stages can be:

• After the official summary date, proponents have a maximum of 180 days to circulate petitions and file signatures ¹.
• An initiative statute needs signatures equal to at least 5% of the votes cast for governor at the last gubernatorial election, and an initiative constitutional amendment needs 8%; for 2026 the Secretary of State lists 546,651 and 874,641 signatures respectively ^{1 4}.
• Once proponents certify that they have 25% of the required signatures, each house of the Legislature must hold joint public hearings, but "the Legislature cannot amend the proposed initiative measure or prevent it from appearing on the ballot" ¹.
• A measure qualifies on the 131st day before a statewide general election, and proponents may withdraw it at any time before it qualifies ¹.
• Since July 2011, statewide initiatives in California appear only on statewide general election ballots ⁴.

That is one state's process. Others differ, and many proposals never go to voters at all. The framework uses a general status model that can map onto each jurisdiction's actual stages:

Each status carries a date and source. Opinions are always shown with the version and status they answered. See Participation over time.

Evidence

• Primary legal and election guidance: California's initiative procedures, including title and summary, fiscal estimate, single-subject rule, circulation period, signature thresholds, legislative hearings, and qualification deadline ^{1 4}. The guide states that it "does not have the force and effect of law"; the underlying statutes and constitution control.
• Comparative overview: availability of initiative and referendum across U.S. states ².
• Comparative example: Swiss federal initiative and referendum requirements ³.
• Design proposal: the framework's required proposal fields and status model. These are design choices, not established standards.

Competing interpretations

"Asking for costs and authority is a barrier to participation." Some argue that requiring this detail makes participation harder and favors professionals and organized groups who can supply it. The framework answers that the burden belongs on proposal authors and reviewers, not on participants, and that missing information should be shown as missing.

"Fiscal estimates are political." Estimates can be contested, and the choice of assumptions can favor one side. The framework cites the source and date of each estimate, shows ranges where available, and does not present any estimate as settled fact.

"Direct democracy bypasses deliberation." Critics of initiative processes argue that ballot measures compress complex questions into yes or no. Supporters argue they let voters act when legislatures will not. The framework takes no side on whether initiatives are good policy; it only requires that any proposal, whatever its route, be described precisely. See Participation between elections and Cases.

Application to this framework

• Issues are prioritized. Proposals receive opinions. Ballot items receive votes. The three are labeled differently everywhere.
• Every proposal shows its adopting authority, cost estimate with source, funding source, dependencies, and dated status.
• Unknown fields are shown as unknown, never filled with guesses.
• Opinions are tied to the proposal version and status they answered.
• The framework never says a proposal "will" be on a ballot unless it has qualified under the responsible authority's rules.

Unresolved questions

• Who verifies the authority, cost, and dependency fields for each proposal, and what happens when they are disputed? See Governance.
• How should competing cost estimates for the same proposal be displayed?
• How should a proposal that spans several jurisdictions be represented?
• When a proposal is amended, should earlier opinions be shown, hidden, or archived?
• Should people be able to propose items, and what review would prevent the list from being flooded?