Madison's conspiracy

• Article
• Comments ()
• Click-2-Listen
• Videos

By

Democrats keep trying to persuade pundits and the media that they have a formula to capture the House in 2006, but a Madisonian principle keeps getting in the way. National polls and political slogans developed in Washington rarely have much impact on 435 locally apportioned congressional races. Incumbent re-election rates, as well as a historically low number of open seats this year, might leave some Democrats wondering if James Madison's concept of local elections for Congress was really a Republican conspiracy. Deconstructing the way to win the House for the Democrats is pretty straightforward — win a net 15 seats — through some combination of beating Republican incumbents or winning open seats previously held by the GOP. But right now Democrats' majority math looks a little fuzzy.

Incumbent re-election rates provide the first peak at a very steep climb. In 1974, Yale University political scientist David Mayhew highlighted growing incumbent safety in a seminal article titled: "Congressional Elections: The Case of the Vanishing Marginals."

The long-term trends Mr. Mayhew pointed out are striking. He documented that in 1948, most congressional districts in America fell into the competitive range (won by less than 55 percent). But over the next quarter of a century the power of incumbency surged. By 1972, Mr. Mayhew notes, 90 percent of incumbents were considered safe.

And while it appears incumbent re-election could not go much higher, in the last few years it has — to more than 95 percent. Between 1946 and 1996, there were only three elections (1968, 1986 and 1988) where the number of incumbents defeated was in single digits. So no matter how you feel about competitive congressional races, they look like endangered species. In the last four cycles, 1998, 2000, 2002 and 2004, fewer than 10 incumbents lost in each cycle. And in 2004, only seven incumbents were defeated in general elections, nearly a 98 percent re-election rate.

If predicting another successful year for incumbents appears like the safe bet, is there another path to Democratic success?

Open seats potentially provide that avenue. But here a little less publicized — yet equally ominous — trend for Democrats is developing. Political professionals know it's easier for a congressional district to switch from Republican to Democrat or vice versa in an open seat, compared to beating an incumbent. That is why party leaders, like National Republican Congressional Committee chair Tom Reynolds of New York and his Democratic counterpart, Rep. Rahm Emanuel of Illinois, have worked so hard to keep down the number of retirements this cycle.

Since 1954 the number of open seats averaged about 40 per election. But in the 2006 cycle, while there still may be some additional late announcements, the current number is only a little over half that (23 total — 15 Republicans, 7 Democrats, 1 independent). And only 10 of the total look truly competitive.

Open seats do produce more turnover than incumbent defeats. But a closer look at the numbers also yields little hope for Democrats. Analyzing all open seats that changed party between 1954 and 2004, the average number of party switches during that 50-year period totaled only 11 seats per election. And as with incumbents, parties are also gaining a firmer grip on holding open seats. For example, in 2004 there were 31 open seats, but only five switched parties.

Democrats might take heart this year, however, because they have to protect just about half the number of open seats as Republicans. Yet this is a pattern Republicans have consistently faced. As University of California, Berkley political scientist Michael Murakami wrote in a 2005 paper presented at the American Political Science Association, except for the 104th Congress, when more Democrats retired following the Republican takeover, between 1973 and 2004, Republicans consistently had more retirements (and hence open seats) than Democrats. 2006 simply continues that pattern. And Republicans have apparently adapted without much electoral consequence.

Attempts by Democrats to nationalize the 2006 election will no doubt continue. However, structural features like incumbent re-election rates and open seats producing less change make Madison's concept of local elections for members of Congress look like a Republican plot. But few recall Madison's original "Virginia Plan" called on the House to also chose the members of the Senate. Is it too late for that idea?