November 1, 2006

Voter Turnout and Congressional Change

by Rhodes Cook

In recent decades, there have been three basic ways that turnout has worked to produce the sort of “big wave” midterm election results that the Democrats are hoping for next week.

First is the “one-party surge,” where one of the parties significantly increases its vote from the previous midterm while the other party’s vote remains essentially unchanged. That is what happened in 1994, when Republicans won control of both houses of Congress. The nationwide GOP congressional vote spiked by more than 9 million votes from 1990, the biggest increase in one party’s vote from one midterm to another since World War II. Meanwhile, the Democratic total declined by 400,000 votes.

Another way to produce a big change in Congress is a “one-party collapse,” where a huge number of voters from one of the parties simply sit out the election. That is what happened in 1974, when the dispiriting backdrop of the Watergate scandal led to a nearly 3-million vote falloff in the Republican House vote from 1970. The Democratic vote grew by barely 1 million. But the GOP drop off was so severe that it cost the Republicans nearly 50 House seats.

A third way to effect considerable change in Congress is what might be called “unequal gains,” where both parties add votes from the previous midterm but one party gains far more than the other. That happened in 1982, President Reagan’s first midterm, when the Republican congressional vote grew by more than 3 million from 1978, but the Democratic tally swelled by more than 6 million. The result: a gain of roughly two dozen House seats for the Democrats.

Figure

Who Votes Determines Who Wins

The common denominator in all three of these elections is that the president’s party took a beating each time. That has been the historic norm, but is not an inviolable rule of midterm elections. In the last two midterms, in 1998 and 2002, the president’s party actually gained a handful of House seats.

Still, one thing is certain. The turnout Nov. 7 will be much lower than the record-high 122 million ballots cast in the 2004 presidential election. Over the last quarter century, midterm turnouts have ranged from 64% to 74% the size of the previous presidential election. Using that range as a guide, it would mean that the turnout this year would likely fall between 78 million and 90 million, translating into either a modest gain over the 73 million votes cast in the 2002 congressional elections or a very dramatic increase.

Figure

A one-party surge in the Democratic vote in 2006 or a one-party collapse in the Republican vote is unlikely, given the GOP’s recent success at voter mobilization. More likely is an increased vote for both parties, with the Republicans hoping to keep the nationwide House tally as close to 50-50 as possible. That is what happened from 1996 through 2000, as the nationwide congressional vote between the two parties was virtually even each time and the GOP retained control of the House.

The GOP’s ultimate ace in the hole is that they approach Nov. 7 as the reigning champions of voter turnout. In 2002, the Republican House vote increased by more than 5 million from 1998, compared with a Democratic gain of barely 2 million. In 2004, President Bush gained more than 11 million votes from 2000, compared with an increase for Democrat John Kerry of barely 8 million over Al Gore’s tally four years earlier.

Figure

GOP strategists are hopeful that the twin ‘Ts’ of terror and taxes, with a dash of gay marriage, will continue to motivate Republican voters this year. But the signs throughout 2006 have pointed to a much more difficult election year for the Republicans than 2002 or 2004.

As poll numbers for the president and the Republican Congress have remained low and stagnant, the targets of opportunity for the Democrats have steadily expanded. A playing field that two years ago featured barely 30 sub-55% House winners has now grown to close to 90 seats, the bulk of them held by Republicans. And it is easy to count at least seven Republican-held Senate seats that could fall to the Democrats.

The Growing Power of Independents

Democrats are not only buttressed by a party base that appears to have remained quite energized since 2004, but also by an increasing flow of independent voters to the Democratic side. In recent presidential and congressional elections, independents have comprised roughly a quarter of the vote and divided about evenly between the two parties. But a variety of recent polls have shown independents breaking decisively this year for the Democrats.

It is a trend that may have more than short-term significance, as independents have emerged as the growth stock in the electorate, while formal allegiance to the Democrats has declined and the proportion of Republicans has remained static.

At least that is the case in the 27 states around the country that historically register voters by party. Since the Republicans took control of Congress in 1994, the proportion of registered Democrats in these states has declined from 48% to 42%, the Republican share has dropped from 34% to 33%, while the proportion of voters signing up as independents (or with third parties, as a comparatively small number do) has jumped from 18% to 26%.

Independents now have a registration advantage in seven states – four in New England (Connecticut, Maine, Massachusetts and New Hampshire), plus Alaska, Iowa and New Jersey. In a number of other states across the Sun Belt, the ranks of the independents has more than doubled over the last dozen years – from 13% to 27% of all registered voters in Arizona, from 9% to 22% in Florida, from 10% to 21% in Louisiana, and from 8% to 18% in New Mexico. In all 27 party-registration states, the proportion of independent voters has increased since 1994.

To be sure, surveys have shown that the number of “pure independents” is 10% of the electorate or less, while most of the other voters that place themselves in the ranks of the unaffiliated actually lean Democratic or Republican. Still, no matter what their degree of independence, these voters are not party loyalists, the “my party right or wrong” type, otherwise they would have registered in a partisan fashion to begin with. In short, the broad spectrum of independents cannot be considered a reliable part of either party’s base and needs to be courted on an election by election basis.

Rhodes Cook is a non-partisan election analyst who publishes a political newsletter. Read the full article at rhodescook.com.