Twelve years ago, I was the first in the national press to write that the Republicans had a serious chance to win a majority of seats in the House. That article appeared in the issue of U.S. News that hit the newsstands on July 11, 1994, less than four months before the election.
That's how late it was in the cycle before anyone except Newt Gingrich and his acolytes took seriously the possibility that the Republicans would win control for the first time in 40 years.
In this cycle, many reporters have been contemplating the possibility that Democrats will take the House back this November. That's partly because most reporters are Democrats and find that result congenial. More importantly, Democrats can take control with a net gain of only 15 seats this year, while Republicans needed 40 in 1994 (and got 52). It's always easier to see how a party can gain 15 seats than 40 -- although 1994 was the only time in the past 20 years that any party gained more than 10.
Democrats' chances of taking those 15 seats are not very good -- if the voting patterns and political contours that have held steady since the 1995-96 budget showdown continue to prevail. Ordinarily in a decade we see a shift in these patterns. Some geographic regions or demographic groups move to one party or the other, or the whole electorate does.
But that hasn't happened in the past 10 years. In the five House elections starting in 1996, Republicans have won between 49 percent and 51 percent of the popular votes, Democrats between 46 percent and 48.5 percent of the popular votes. Nor have regional patterns changed much. From 1990 to 1996, the nation's largest metro areas became more Democratic, while rural areas and the South became more Republican.
Since then, things have stayed about the same. And this is regardless of whatever problems were facing party leaders like Bill Clinton, Gingrich and George W. Bush.
The redistricting that followed the 2000 census was based on those same voting patterns. That's why so many safe seats resulted from Republican gerrymanders in Florida, Pennsylvania, Michigan and Texas; Democratic gerrymanders in North Carolina and Maryland; and bipartisan incumbent-protection gerrymanders in New York, California, Illinois and Ohio. If the political contours should shift, as they did in the 1970s, 1980s and 1990s, then some seats designed to be safe will become marginal, and some will shift to the other side.
That's what Democrats hope is happening this year.
Pollster Stanley Greenberg says his latest Democracy Corps poll shows Republican support falling sharply in the Deep South, in rural areas and among downscale men -- groups among which Republicans have had big leads. Such a change wouldn't affect many House races, because these groups are concentrated in districts that are very heavily Republican. But it could put another dozen or so Republican seats in play, over and above the dozen or so where Democrats are making strong challenges (Republicans are making strong challenges in about half a dozen Democratic seats).
In years when voters have shifted sharply to one party -- Democrats in 1974, Republicans in 1994 -- the winning parties captured only about half the seats they targeted.
So even if the field of contested seats expands as Greenberg suggests, Democrats could take the House only if they picked off half their targets, while defending every one of their own contested seats. But few seats are captured without strong challenger candidates, and while Democratic recruiting has had some successes, it hasn't produced serious challengers in all these seats. Democrats have a chance to win the House, but it's far from a sure thing.
Of course, not all the factors have played out. At this point in the 1994 cycle, the Clinton healthcare plan had not yet collapsed, the Democrats had not yet embraced gun control and the Republicans had not yet rolled out their Contract With America. Will rural voters, however, cross with Bush and vote to install as speaker of the House a San Francisco Democrat?
Finally, the polls, whatever their bad news for Republicans, offer few clues about who's actually coming out to vote. In 2004, Republicans won because they did a better job turning out their party base than the Democrats did. They expanded the electorate and have a bigger reservoir of voters to draw on. My guess is that turnout, more than voter shifts, will determine who wins in 2006.