An amazing mystery about recent voucher referendums in Oregon, Colorado and California is that vouchers, which seemed to be popular in the beginning, were defeated by margins of at least two-to-one. How did this happen?

Many people are attracted by the idea of vouchers--especially when the proposal goes by an appealing name such as "parental choice" or "market schools." However, when they consider specific voucher proposals, there are usually basic elements that they find totally unacceptable. It's a little like the people who strongly favor the idea of a balanced budget but oppose all the ways of achieving a balanced budget--like cutting programs or raising taxes.

One basic element in a voucher plan is who gets the vouchers. You can set it up so all school-age children are eligible, including those in private schools. Or you can decide to give vouchers only to youngsters in public schools. When vouchers are restricted to public school youngsters, private school parents strongly oppose the scheme. On the other hand, if kids already going to private school get vouchers, it takes lots of additional money, and where does that come from? Do you cut spending for public education? Raise taxes? Most people do not want to hurt the public schools. But they don't want to pay more taxes either.

Another consideration is where the voucher can be used. Is it limited to schools that accept it as full tuition? This is a good thing for youngsters whose parents don't have any money to add to a voucher. Or is the voucher available for any school, no matter what its fees?

This means that wealthy parents are free to use the voucher to replace part of what they now spend, so taxpayers are helping to subsidize the education of youngsters whose parents already foot big bills themselves. On the other hand, poor kids may not find any school where the tuition corresponds with the voucher, and their choice is likely to be very limited. Allowing everybody to have the voucher exacerbates the inequalities between rich and poor. However, limiting the voucher is likely to annoy a large segment of voters who want to spend a little less on their children's education or get something a little better.

A third major consideration is the basis on which students are admitted to voucher schools. Advocates have identified vouchers with parental choice. In fact, the choice has always been up to the private schools. With vouchers, you can continue allowing private schools to choose their students, but that breaks the promise of parental choice, currently one of vouchers' big selling points. Or you can have a system that admits kids on a first-come, first-served basis, by lottery or on the basis of need. An arrangement that takes the choice out of the school's hands is the fairest, but it destroys what makes private schools attractive to many parents. They don't want the youngster with the greatest need sitting next to their child; they want a kid who fits in.

There is a built-in problem for people who try to write a voucher proposal that will pass a referendum. It resembles the problem Democrats and Republicans face in the primaries and the general election. In the primaries, the left wing and the right wing of Democratic and Republican parties, respectively, are very influential because they turn out disproportionate numbers of their people. So if you're George Bush, you have to sound a lot like Pat Buchanan to get the nomination, and the same is true for a Democratic candidate. The problem is, after you've won the primary, you have to convince the general public, which is usually moderate, that you're not some kind of nut or you will lose the election.

The history of vouchers in California illustrates this very well. Proposition 174 was designed to satisfy very conservative voters, and those people worked hard to get it on the ballot. But it did not appeal to people in the middle, so it was defeated. In contrast, John Coons, a professor at Berkeley, wrote a voucher proposal a few years ago that would have helped the youngsters who needed it most. Perhaps it would have passed in a referendum, but he couldn't get enough signatures to get it on the ballot. If a voucher proposal is to make it, it has to appeal to extremely conservative voters because they are the ones who will work to get the referendum before the public. But then you also have to make it acceptable to the general public.

How will voucher strategists proceed now that they've failed in the California referendum? Will they forget about taking it to the people and instead try to sneak it through a state legislature?