election that was too close to call
James E. Campbell
Professor, UB Department of Political Science
more than five weeks after election day, the presidential outcome
remained undecided. Until the U.S. Supreme Court issued its ruling
on December 12, and Al Gore conceded the next day, no one could
know for certain the identity of the nation’s 43rd president.
Earlier, as we wallowed waist-deep in court rulings and ballot
chad, awaiting the outcome and pondering whether votes in Florida
were being counted or stolen, one thing was known for certain:
Based on the number of votes that could have changed who won
the presidency, this was the closest election since popular
voting became widespread in 1828. It was much closer than the
Kennedy-Nixon race of 1960 and even closer than the disputed
Hayes-Tilden election of 1876. While George W. Bush was the
electoral college winner, Gore received 50.2 percent of votes
cast nationally for the major candidates.
Why was the election so close? Presidential elections are normally
quite competitive, but two conditions set the stage for a particularly
close election this year. First, there was no incumbent running
and elections are closer without the galvanizing presence of
one. Nearly forty percent of presidential elections since 1828
without an incumbent in the race were won with less than 51.5
percent of the two-party vote. The odds of an election this
close are nearly five times greater without an incumbent in
the race than with one.
Partisanship also favored a close finish. The number of Democrats
and Republicans in the electorate are now more evenly balanced
than they have been for many years. From 1952 to 1980, Democrats
outnumbered Republicans by more than 15 percentage points. Since
then, their lead has eroded to barely three points. With the
parties at parity, neither candidate had an audience of voters
either more hostile or more receptive than did the other.
Beyond these two factors, however, the campaign took unusual
twists and turns that, in the end, brought us to a near dead
heat. Take the economy. According to Gallup polls, the number
of Americans saying that the economy was good or excellent more
than doubled from 1992 to 1996 and more than doubled again in
2000. There was no doubt as to how most voters would answer
the question: “Are you better off today than you were four,
or even eight, years ago?” Voters normally give credit to the
in-party candidate when the economy is strong. A successor might
not receive the same credit as the incumbent, but he could take
a big share of it. Largely because of this, all of the
forecasting models predicted that Gore would win the popular vote
and all overestimated the size of the vote for Gore. My own model,
based on the Labor Day preference polls and the second quarter
growth rate in the economy, predicted that Gore would receive
about 52.8 percent of the two-party vote.
While there are several possible explanations for why the models
overpredicted the vote for Gore, the most plausible is Gore’s
campaign. In an effort to steer clear of fallout from the Clinton
scandals, Gore ran a prospective rather than the anticipated
retrospective campaign. He said as much in his acceptance speech
at the Democratic convention: “This election is not an award
for past performance. I’m not asking you to vote for me on the
basis of the economy we have.” This is perhaps the oddest twist
of the campaign: in the midst of a strong economy, an in-party
candidate with a strong economy who refused to make it the centerpiece
of his campaign.
Despite the fact that Gore had a much easier time than Bush
in winning his party’s nomination, Republicans were more united
and this was reflected in Bush’s consistent poll lead over Gore
throughout the summer. Apparently, eight years of dealing with
President Clinton convinced Republicans to set aside their internal
party squabbling and get behind a candidate whom they thought
could retake the presidency.
The race turned yet again after the Democratic convention.
Candidates who trail in the polls usually get a bigger bump
from their conventions and Gore made good use of his to rally
his base. With Democrats energized, Gore pulled into a modest
lead in the polls.
Although voters saw the October debates as a draw (with Gore
winning the first, Bush the second and a tie in the third),
Bush regained the poll lead. The debates dispelled or weakened
the impression of Bush as not being smart enough to be president.
A caricature of Bush that had been promoted could not survive
the reality check of the debate for many voters.
In the final week of the campaign, with late-deciding voters
coming back to their parties and splitting more evenly between
Gore and Bush, the Bush lead evaporated. On election day, all
of the polls said the election was too close to call. No one
thought that we would be saying the same thing more than a month
later and that the twists and turns of campaign 2000 would continue,
to the amazement of everyone and to the dismay of many.
James E. Campbell is the author of The American Campaign: U.S. Presidential
Campaigns and the National Vote (Texas A&M University Press, 2000).
His theory of predictable campaigns is based on a decade of research
on electoral history over the past 130 years. He is often called upon
by the national media, including the New York Times, the Washington
Post and others, for his analysis of campaigns and elections.