Every presidential election year there is a chorus of complaints about exit polls and the television networks' tendency to project final election results while some of us are still voting. With the Iowa caucuses on Monday and the New Hampshire primary next week, the music should be starting soon, fortissimo.
Above the din it is reassuring to hear some good common sense on the issue of exit polls from the U.S. Court of Appeals for the 9th Circuit. Last week a three-judge panel unanimously struck down a 1983 Washington state law prohibiting exit polling within 300 feet of any polling place.
Washington state officials had contended that the law was designed to preserve order at the polls. But the appellate court, noting that there were already laws on the books banning disruptions at the polls, called the restriction on exit polling a ruse--a clumsy attempt to keep pollsters so far away that they could not speak to voters and to deprive the news media of an essential reporting tool. Under the First Amendment, the court held, that is impermissible.
As Judge Warren J. Ferguson wrote, the state law didn't even make sense on its own terms. After all, what upsets some Washingtonians and other Westerners is not so much the exit polls done in their own states but the networks' Election Night projections of the results from states farther east. The Washington law wouldn't stop the TV reports and couldn't change the fact, dictated by time zones, that the polls close first in the East and the returns soon start pouring in.
And yet, even if Washington couldn't devise a solution that would pass constitutional muster, we sympathize with the state's concern, shared by many Westerners, that the electoral process somehow is being short-circuited.
No one has ever established conclusively that exit polls and early television projections change the outcome of races. But after the 1980 election, when President Jimmy Carter was persuaded by exit polls to concede to Ronald Reagan at 6:15 p.m. PST while West Coast polls were still open, the University of Michigan conducted an authoritative study concluding that up to 11% of voters in some precincts didn't vote because of Carter's concession. The main horse race was over, so they stayed home--a decision that may have changed the outcome in several local races.
Americans annoyed by what they see as network meddling in elections have offered dozens of ideas, most of them terrible, about what to do about exit polls and projections. Former Sen. S. I. Hayakawa wanted to slap a $1,000 fine on anyone who made public election results before the polls closed. Columnist Mike Royko urged voters to lie to exit pollsters about how they had voted. And, in 1984, both houses of Congress passed nonbinding resolutions urging the networks to exercise self-restraint in calling elections.
The best idea that we have heard comes from Reps. William M. Thomas (R-Bakersfield) and Al Swift (D-Wash.). Their bill (HR 435), approved by the House and now before the Senate Rules Committee, would impose a uniform closing time on polls across the country. Except in Alaska and Hawaii, all polling places would close at 9 p.m. EST--8 p.m. in the Midwest, 7 p.m. in the mountain states.
To give Westerners a chance to vote after work, daylight-saving time would be extended for two extra weeks in the West every presidential election year--so polls on the Pacific Coast would close at 7 p.m. PDT. That extra hour is important: In Los Angeles County about 14% of registered voters consistently vote between 6 p.m. and 8 p.m., the current closing time.
The uniform poll-closing bill may not be perfect. The airlines, for example, worry that the daylight-saving feature would snarl their schedules. But, as Thomas has observed, the proposal has the virtues of respecting the First Amendment, of allowing the media to report what they know as soon as they know it and of offering "the most minimal intrusion into the electoral process."