The following is a frank admission, and we hope it will not be held against us: More than once in the past, we've entered a voting booth only to discover that there was some kind of ballot measure we had never even heard of. We would try to figure out what the proposal meant on the spot, but this can be futile. Sometimes it's tough even figuring out whether a yes vote means we're voting for or against something.

Just because we're sheepish on this point doesn't mean we can't find fault with others. Over the past election season, as we followed a number of significant referenda and initiatives in several states, we were startled at the lack of good information being published to help voters make intelligent decisions. The most obvious problem cropped up in the dollar costs associated with various proposals.

For example, there was a measure in Arkansas that would have eliminated the sales tax on food. The proposal was a bit vague about what was actually being exempted. Prepared foods? Sodas? Before they turned the measure down in November, Arkansas voters saw newspaper estimates of lost revenue that often ranged from $150 million to over $500 million. And it occasionally got worse. In a September 25th article in the Arkansas Democrat-Gazette, one proponent of the amendment indicated that the measure would cost the state only a piddling $1 million to $8 million. The article didn't challenge this comment; it just published it as one more confusing fact with which voters would have to grapple.

In Michigan, a constitutional amendment was proposed to provide the option of binding arbitration to all state employees over "wages, hours, pensions and other terms and conditions of employment." Representatives of the governor's office, which opposed it, put the price tag at $500 million over three years. The state's Chamber of Commerce, which was on the same side, argued that it would cost about $180 million. We chatted with representatives of both organizations. It turns out the Chamber's number was an estimate of the additional salary costs associated with the proposal. The governor's people were also adding in their estimates of the impact of extending bargaining to items such as pensions and conditions of employment. For all we can tell, both figures may have been right. But voters had to be mystified by the range, which went unexplained in clips we saw.

In Florida, there was a proposal to limit class size. Supporters of the measure--which narrowly passed--put the cost at about $8 billion through 2010. Opponents, including Governor Jeb Bush, were inclined to use a $27 billion figure. Both figures appeared in many news reports. We certainly understand that it's tricky coming up with accurate projections for programs such as this. But a $19 billion range eludes any rational explanation. If a car dealer told you he'd happily sell you that cool red number on the lot, only he couldn't tell you whether the price was $8,000 or $27,000, would you be inclined to stick around?

Serendipitously, the same November ballot with the class-size measure also had an amendment requiring a clear price tag for any citizen- proposed amendment. It passed and will go into effect next spring. Representative Randy Johnson, who proposed adding the amendment to the ballot, said, "When we make decisions in family budgets, we don't ignore how much things cost."

That seems like a natural solution. It's similar to laws in most states requiring fiscal notes be attached to legislation before lawmakers vote. Why should citizens be given less information when they're effectively doing the same job?

Florida was the eighth state to pass such a requirement. The other seven were California, Mississippi, Missouri, Montana, Oregon, Utah and Washington. The National Conference of State Legislatures has gotten on the case, and its Initiative and Referendum Task Force recently recommended that states should require the drafting of a fiscal impact statement on every proposal offered to the voters.

Curiously, many of the laws that already require this kind of information--including the new one in Florida--demand fiscal notes for initiatives that are placed on the ballot by citizens but have no parallel condition for ballot measures that are advanced by the legislature. "Lawmakers just exclude their own process," Dane Waters, president of the Initiative and Referendum Institute, told us. "And that's been bothersome to us as an organization. We believe what's good for the goose is good for the gander."

That certainly seems sensible. And Waters makes another important point: He believes that laws requiring fiscal notes on ballot measures should also create a process for allowing challenges to the number provided by a bill's sponsor. Otherwise, everything would be portrayed as cheaper than cheap.

But assuming that kind of process, it seems to us that governments can only benefit when taxpayers know exactly what they're voting for, as well as how much it costs. Then the big trick is getting taxpayers such as us to actually read that stuff.