The Ballot Measure Has Become the Legislature

The initiative process was built as a release valve. When a legislature became captured or complacent, citizens could put a question directly to the electorate and settle it themselves.

It works, in the narrow sense that measures pass and become law. What it does not do is produce good statutory text, because a ballot measure cannot be amended once it is circulated. Whatever drafting error made it into the petition is the law the voters get.

The signature economy

Qualifying a measure in a large state now costs several million dollars, nearly all of it spent on paid circulators. That price is a filter. It admits campaigns funded by industry associations, unions, and a small number of wealthy individuals, and it excludes almost everyone else.

The framing does not change with the funding. Every campaign presents itself as a grassroots revolt against an unresponsive capitol, because that is the framing that polls best, and the disclosure that would complicate the story appears in a filing nobody reads.

A process designed to route around concentrated money has become one of the most reliable ways to deploy it.

What happens after passage

The measure passes and then meets the rest of the legal system. Agencies have to write implementing rules for text that was drafted to be persuasive rather than administrable. Courts have to reconcile it with existing statutes that the drafters may not have read.

Legislatures, meanwhile, often hold the power to amend or delay, sometimes requiring a supermajority, sometimes not. The fights that follow a successful measure can run longer than the campaign did, and they happen without the attention that the campaign attracted.

The case for keeping it anyway

None of this argues for abolition. On several questions where legislatures were genuinely immovable for decades, initiatives produced changes that large majorities wanted and could not otherwise obtain.

The reforms that would help are modest and boring. A mandatory legislative review window before signatures circulate, so drafting errors can be caught. Plain language summaries written by a neutral office. Real time disclosure of who is paying for circulation. None of these limit what voters may decide. They only make it likelier that what passes says what its supporters thought it said.

The original design assumed a specific failure and was well matched to it. A legislature captured on one question could be routed around by citizens who cared enough to organise, and the cost of organising was mostly time.

That cost is now money, and the change is not subtle. Qualifying a measure in a large state requires paid circulators at a scale that only a funded campaign can sustain.

The election code sets the signature threshold and the window, and the two together determine the price. Neither was chosen with a professional signature industry in mind, because there was not one.

The framing has not adjusted. Every campaign presents itself as a revolt against an unresponsive capitol, because that is the story the mechanism was built for and it remains the most effective way to sell a measure.

The drafting problem is the part with the longest tail. A measure cannot be amended once it is circulating, so an error found in month two is on the ballot in month nine and in the statute book after that.

The implementing statutes then have to accommodate text written to persuade rather than to operate, and the accommodation is done by agencies with no authority to fix what is actually wrong.

Litigation follows predictably. A provision that is unclear gets construed by a court, which produces a meaning nobody voted for and which the drafters may not have intended either.

How states structure the process varies enormously, and the variation is instructive. Some require a legislative review window before circulation, and the measures that emerge from those states are noticeably better drafted.

That reform is the obvious one and it is resisted for a reason worth stating fairly. A review window gives the legislature a chance to interfere with a mechanism designed to bypass it.

The answer is to make review advisory and mandatory to receive rather than to accept, which preserves the bypass while removing the drafting errors that currently survive into law.

The administrators who run it carry the consequences either way, and they are the constituency least represented in the argument about how the process should work.

None of this argues for abolition, and the record does not support it. On several questions where legislatures were immovable for decades, initiatives produced changes that were durable and popular, which is the case the mechanism has to be judged on.