Commentary
I Vote for High-Tech Balloting
John Steele Gordon / Wall Street Journal 8nov00

By John Steele Gordon, author of "The Great Game: The Emergence of Wall Street as a World Power, 1653-2000," out this month in paperback from Touchstone.

When Americans went to the polls Tuesday, we chose the leader of the most technologically advanced country on earth. We have more automobiles, computers, telephones, televisions, VCRs, Cuisinarts, and heaven knows what else, per capita than any other country. Yet when we voted -- the central act, the communion if you will, of democracy itself -- we Americans mostly used technology so primitive that it was available 100 years ago, in a few cases 200 years ago.

The paper ballot, still in use in a few places, dates back centuries and was the method used to elect George Washington. Not much changed for another 100 years, except the scale and scope of voter fraud.

Inventors began to look for ways to counter fraud and, as important, make it easier to cast and count votes. The very first patent awarded to Thomas Edison, in 1868 when he was 29, was for a device that would allow congressmen and senators to vote on motions by simply pushing the appropriate button. Voice votes that often took 45 minutes could be replaced with a nearly instant tally. But Congress wanted no part of Edison's bright idea, much to the young inventor's chagrin. Its members preferred the old method, which allowed pressure to be put on late-voting legislators.

In the 1890s machines for voting by the public were developed. In 1891 New York made the Myers Automatic Booth legal for use in the state and in 1892 it was first used in an election, in upstate Lockport. These early machines were quite primitive, of course. Basically they were simply a series of levers mechanically linked to counters. Before voting began, officials would set the counters to zero, and after the polls closed they would read off the results and send them by messenger to a central place where they were tallied with results from other precincts.

Needless to say, it didn't take long for those more interested in winning than in democracy to figure out how to subvert the new technology. A paper tape, punched to record each vote, made fraud more difficult but added to the already considerable mechanical complexity of the machines. By the 1920s, when mechanical voting booths finally began to widely replace the paper ballot, they were so complex that Rube Goldberg would have been embarrassed to have designed them.

But then a strange thing happened: While new technology was applied to almost everything else in American life as soon as it became available, voting machines stayed frozen in the technology of the mid-1920s. In the 75 years since, we have gone from Charles Lindbergh crossing the Atlantic to Voyager II crossing the solar system, from the Model A to the sport utility vehicle.

But we still vote using machines the size of large refrigerators (they weigh about 800 pounds) and far more mechanically complex. Eligibility to vote is still determined by officials looking in large books of registered voters delivered from the local board of elections. The results have to be transmitted manually to a central tabulating center and only then released to the news media.

Why?

The answer is simple. Just as competitive capitalism forces innovation, so monopoly discourages it. That's precisely why innovation in telephone service exploded (and prices crashed) only after AT&T was broken up in 1984. And government is the largest monopoly of all. Most boards of election are perfectly happy with the status quo. Even when some new technology has crept in, it is a half-way measure at best. In some jurisdictions, computerized voting machines have replaced the old mechanical ones, but that is all they have done. The ladies with the great books are still there, looking up every voter. The results are still officially delivered by hand to the various boards of election.

And yet the technology of a truly modern (and, at least for the moment, fraud-proof) voting system is already well-developed and ready to go. It is not the Internet; the problem of how to prevent fraud on such a wide-open communications platform has not been solved by any means. Instead it is a technology that is familiar to everyone and sufficiently secure that banks are willing to dispense millions by means of it every day: the ATM.

There is not a single technological reason why a voter on election day should not be able to walk into any convenient polling place, insert his credit-card-like registration card and his PIN number, answer a series of questions on a screen ("Please vote for one of the following candidates for president"), review his choices, and press a button to register his votes. The votes would be transmitted instantly to the central computer and at the close of polling, results would be instantly known. If a nationwide 24-hour election period were adopted, people in the eastern United States could still get a good night's sleep on election night.

But while there are no technological reasons such a system could not be well on its way to being implemented by the next presidential election, the vast inertia inherent in government -- not to mention the self-interest of politicians comfortable with the status quo for reasons honorable and less so -- makes that extremely unlikely. So while I can dictate this article to a voice-recognizing software program instead of typing it on a manual typewriter, I will still have to vote as my great grandparents did.

One could hardly ask for a better example of the superiority of private enterprise over government when it comes to producing goods and services.