There’s No Benefit in Nationwide Polling Hours

Procrustes, a mythological figure, cared greatly for the comfort of his guests when they slept. But rather than adjust the bed to his visitors, he did the opposite. Guests who were too short were stretched; those who were too tall had their feet lopped off.

Rube Goldberg was a newspaper cartoonist best known for “inventions” that accomplished the simplest tasks by absurdly complex and indirect means.

A bill passed last week by the House of Representatives, requiring that the polls for presidential elections close at the same moment in all states except Alaska and Hawaii, is jointly inspired by Procrustes and Rube Goldberg.

Because of time differences, the polls would have to close late at night in the East and early in the evening in the West. That’s the Procrustean part of the bill.

The bill’s sponsors didn’t want the disparity to be too great, so they added a Rube Goldberg feature, requiring that California and the other states in the Pacific time zone extend daylight saving time for an extra two weeks. For 50 weeks of the year, Californians will be the familiar three hours behind Easterners. But for two weeks, including Election Day, we shall be two hours behind.

What is the purpose of this proposal? It is a carry-over from the 1980 election, when NBC projected Ronald Reagan as the winner three hours before the polls closed in California. President Jimmy Carter then conceded the election with more than an hour of voting time left on the West Coast.

Supporters of the House bill assume that these events discouraged masses of people from voting. They hope to prevent this from happening again by closing the polls everywhere simultaneously. But their plan is flawed in half a dozen important respects.

First, it is not at all clear that early projections of presidential results have a major effect on voting. The political science journals contain numerous studies of this question. Nearly all of them conclude that early projections either have no effect on turnout, or only a small one.

Even if there is a problem, the bill won’t solve it. Network projections are based not on early vote returns, but on exit polls. If they wish, the networks can make projections and publicize them long before the polls close anywhere. In 1984 the networks voluntarily refrained from doing so. They may or may not display continued restraint in the future, but whatever they do will not be affected by the House bill.

The 1980 election was exceptional in that an expected close contest turned out to be a landslide. In more typical landslide years like 1964 and 1972, anyone with a television or newspaper knows in advance what the result will be. In most cases, the harm that this bill is intended to prevent will have occurred already, from polls published before Election Day.

And if the bill were effective, the result might be perverse. In 1948 and 1968 expected landslides turned out to be close. If the House bill worked, Californians would be deprived of information that might encourage them to vote.

The House bill would require us to close the polls at 7 p.m. instead of 8 p.m., as California law now provides. This is a strange way to go about encouraging people to vote. According to the Los Angeles County Registrar of Voters, between 5% and 9% of the vote typically has been cast during the last hour. In contrast, where the political science studies found that early projections do result in a turnout reduction, it usually has been estimated at only 1% to 3%.

Will some of the people who previously have voted between 7 and 8 p.m. vote earlier if the polls chose at 7? Certainly. But the number of voters lost by the forced early closing of the polls probably will exceed the number of voters who might be gained if the House bill somehow prevents the publicizing of projected results before the polls close.

Furthermore, to the extent that the early closing forces 7-to-8 o’clock voters to cast their ballots earlier, the result will be longer lines at the polls for all. California Sens. Alan Cranston and Pete Wilson, each of whom has announced support for the legislation, might want to contemplate the likelihood that future opponents will warn the public of who voted to lengthen those lines.

Finally, the daylight savings feature will have annoying side effects. Californians are used to the three-hour time difference. How many people will waste an hour in an airport or fail to buy or sell some stock or miss the first quarter of a Rams game (if anyone still watches Rams games) because of confusion over the two-week change in the time difference?

These nuisances are not unbearable, but why should we accept them for the sake of small, speculative voter turnout gains that are likely to be offset by equal or greater turnout losses?

The Senate killed similar legislation in 1986. This time the senators should bury it.