Here We Go Again: An Anti-Incumbent Wave Next Year?

by Stuart Rothenberg November 1, 2007 · 12:05 AM EDT

Some story lines never go away.

This year, once again, there is buzz that 2008 might be an anti-incumbent election that will sweep out sitting House Members of both parties. National Republican Congressional Committee Chairman Tom Cole (Okla.) has been making that case for months, and more than a few journalists and talking heads have picked it up as well.

A little more than a year ago in this space (“An Anti-Incumbent Election? This Year? Of Course Not,” Sept. 14, 2006), I argued that 2006 would be an anti-Republican, not an anti-incumbent, year. I never thought that we’d be hearing the same anti-incumbent argument so soon. It’s like a bad penny that keeps turning up.

According to “Vital Statistics on American Politics, 2007-2008,” over the past 27 elections, dating back to 1954, there have been 11 or 12 partisan blowouts (depending on how you classify 1984), where one party or the other has suffered big losses and the other party had few or no incumbent defeats.

Last year’s election, when 22 GOP House incumbents were defeated without a single Democratic casualty, is a perfect example. The 1960, 1964 and 1996 elections, to cite three cases, produced similar partisan waves.

In eight elections since 1954, there were small, single-digit incumbent losses. They certainly were not “anti-incumbent” elections. In 2000, for example, six incumbents were defeated in November. By any definition, that was a status-quo election, not an anti-incumbent anything.

That leaves seven of the 27 elections that were neither blowouts nor status-quo contests. Were any of them “anti-incumbent” elections? Maybe, but probably not.

Of those, the worst year for incumbents was in 1992, when a total of 24 House incumbents — 16 Democrats and eight Republicans — lost to challengers. Cole, who served as NRCC executive director back then, has cited the ’92 elections as an example of a year when voters directed their anger at incumbents of both parties (and ousted a sitting president).

Twenty-four House incumbents going down to defeat may well qualify as an anti- incumbent election in the abstract, but, alas, it’s more complicated than that. The devil is in the details.

Large numbers of incumbents lost that year because of scandals and redistricting, not because voters across the country were so angry with Capitol Hill or with politicians in general that they simply voted against incumbents, regardless of party. The 1992 CQ Almanac did a wonderful job documenting this in its end-of-the-year rehashing of the election results.

“Voter discontent and redistricting did take a toll on members who sought re-election, but the much-discussed possibility of an Election Day cyclone of anti-incumbent sentiment failed to materialize Nov. 3,” the almanac’s authors wrote.

Some incumbents lost because their districts had been redrawn to include more opposition partisans who voted primarily because of party. Others lost because they bounced checks on the House bank or were under indictment. Some lost because of the top of the ticket. Few, if any, lost merely because they were incumbent officeholders.

So classifying1992 as an “anti-incumbent” election is committing a classic mistake: focusing only on the aggregate numbers and ignoring the individual results.

This isn’t to say that in 1992 incumbents weren’t at all affected by the public’s mood, which held politicians in particularly low repute. Some incumbents saw their numbers slip, and the percentage of House elections won with 60 percent of the major-party vote, according to “Vital Statistics on Congress, 2001-2002,” slid from 88.5 percent in 1988 to 65.6 percent in 1992.

The election with the next largest number of incumbent defeats was 1978, when 19 total incumbents lost in November. That election, however, had even a stronger partisan hue, since 14 Democratic incumbents lost in November to just five Republicans. It was a midterm, anti-Democratic election.

Total incumbent losses in the remaining five cases — the elections of 1956, 1962, 1970, 1976 and 1990 — were neither small nor huge. They were in the low to middle double digits, ranging from 11 to 15 incumbents. Those levels of losses seem below a “wave” to me, but some may disagree.

So how does 2008 fit in to this analysis? Those who believe that an anti-incumbent wave is in the works surely have an argument to make, and they should not be dismissed out of hand.

The past nine major national polls measuring the public’s attitude toward Congress have all produced scary results for Democrats. Each of those surveys found Congress’ job-approval to be below 30 percent, and in each lower than President Bush’s job-approval rating.

Given these poll numbers, and Congress’ inability to deal with big issues, including Iraq and immigration reform, it’s theoretically possible — but extremely unlikely — that some voters will decide that the current Congressional membership is to blame for the nation’s problems and will vote against all incumbents, Republicans and Democrats alike, a year from now.

When voters are confused about whom to blame, they either support incumbents (blaming the other Representatives, not their own) or they fall back on their fundamental partisanship and vote their party. In either case, they don’t vote “against incumbents” merely because of their incumbency.

Democratic voters are angry with Congress now for not taking Bush on strongly enough over the war in Iraq. That explains why Congress’ job ratings are so low. But that anger is likely to dissipate, and it is difficult to imagine Democrats (or even independents) voting against Democratic House incumbents merely to make a point.

That doesn’t mean that incumbency, to some voters and in some districts, won’t be a liability. Unconventional candidates and messages of change could attract voter attention. But a rabidly anti-incumbent election is not as likely as some hope.