By Thomas F. Schaller
Although I have great confidence in the empirical findings of Jon Krasno and Alan Abramowitz regarding the impact of gerrymandering on U.S. House competitiveness, their conclusions do not preclude Democrats from trying to win as many governorships and state legislative seats between now and 2010 in order to exercise maximum possible influence during the next round of redistricting. At that point, in many states a real opportunity for Congressional Black Caucus (CBC) and Congressional Hispanic Caucus (CHC) members will present itself: Will they gamble a bit of electoral security in exchange for a promise of greater party or committee power if and when Democrats recapture the House majority?
Regardless of their race or the racial composition of their districts, far too many Democrats in the U.S. House are representing too-safe districts, a reality which prevents the party from maximizing its House seat share. The “unholy alliance” forged between Republicans and minority Democrats led to the election of more CBC and CHC members to Congress, but also more Republicans. As well, many white Democrats enjoy unusual electoral security. A new alliance between white and minority Democrats must be forged, with the goal of redrawing the 2012 maps to enable Democrats to recapture-or, if already recaptured, retain-a House majority.
To understand the extent of the problem, look at the figure below, which depicts the Partisan Voting Index of all 435 House districts, as published by the Cook Political Report. PVI measures the congressional district performance of presidential candidates during the past two presidential cycles: Positive numbers indicate districts Bush carried in 2000 and 2004, and vice versa for the Kerry/Gore districts. For visual clarity, the Democratic districts were signed positively and then all 435 districts were arrayed, left to right, from the most Democratic district (Jose Serrano’s NY16 = -43.4 percent margin for Kerry) to the most Republican (Chris Cannon’s UT3 = +26.2 margin for Bush).
Notice that, including Serrano’s, there were 31 districts with higher Democratic PVIs than Bush’s 26.2 margin in Cannon’s Utah district. This is troubling. Indeed, Democratic voters are distributed so inefficiently than the 199 Democratic-leaning districts on the left side comprise a greater total area than the 236 Republican districts on the right side. Given that Bush’s combined popular vote from 2000 and 2004 exceeds that of Gore and Kerry combined, what explains the apparent paradox of the right side containing less area than the left? The answer is simple: The bars in the figure are percentages, not absolute vote margins, and only if turnout were identical in every district would the Republican side necessarily be larger.1
Turnout is far from identical, however. Consider the accompanying table, which reports the 2004 turnouts from 20 districts: a set of 10 Democratic districts and another 10 Republican districts with almost the same, opposite-signed PVI scores. The average turnout in the 10 Republican districts was 253,837, compared to just 214,121 in the 10 Democratic districts-about 40,000 fewer votes, or 16 percent lower turnout. These differences cannot be explained by the 2004 presidential contest, because the only swing-state district among the 20 is Democrat Gwendolyn Moore’s 5th District in Wisconsin which, not coincidentally, had the highest turnout of the 20. Despite Moore’s total, the 10 Republican-represented districts still had higher turnout. Notice that all but two of the 10 Democrats (New York’s Carolyn Maloney and California’s Howard Berman) are CBC or CHC members.
Because socioeconomic status affects turnout, the majority-minority districts represented by these mostly black and Hispanic members feature some of the lowest turnouts in the country. Can anyone blame their constituents for not showing up? If they live in a non-swing state with a safe Democratic incumbent running literally, if not virtually unopposed, why bother? Many racial minorities also reside in non-competitive state legislative districts. As my colleague Tyson King-Meadows and I report in our new book, Devolution and Black State Legislators, typically about 90 percent of black state legislators win with at least 60 percent of the vote, with 60 percent winning with at least 90 percent! When and where possible Democrats should try to produce a more competitive set of maps in 2012 which, though still ensuring the election or re-election of minority Democrats, also induce higher turnout among Democratic base voters. To accomplish this, in any state where they exert power over the redistricting process Democrats should duplicate what might be called the Cummings-Wynn model from the 2000 round of redistricting in Maryland.
Prior to 2002, in a state that was overwhelming Democratic in both legislative chambers and had not elected a Republican governor since 1966, Maryland’s eight-member U.S. House delegation was split, four seats for each party. The Democratic governor, Parris Glendening, along with the Democratic Senate President and House Speaker created a new map to tip the state’s split delegation in favor of Democrats, six seats to two. How? African American voters from black Congressman Al Wynn’s Prince George’s-based 4th District were moved into Republican Connie Morella’s 8th District, and white suburban voters from Howard County were moved into black Congressman Elijah Cummings’ 7th District to free up extra Democrats for the 2nd District that Robert Ehrlich had vacated to run for governor. Capturing both seats required Cummings and Wynn to run in slightly less favorable and familiar districts. Yet Cummings and Wynn still won re-election easily.
The lesson of the Cummings-Wynn approach is simple: CBC and CHC Democrats have the surplus voters the party can redistribute elsewhere to create a larger set of competitive districts. This is an asset CBC and CHC members should neither horde nor bargain away. If Nancy Pelosi wants to be the first woman Speaker of the House of Representatives, she should broker a deal with black and Hispanic members of her caucus in which they agree to assume a bit more electoral risk in exchange for two promises: (1) that the Democratic Congressional Campaign Committee will rally behind any black or Hispanic incumbent who faces a serious primary or general election challenge; and (2) that the CBC and CHC will be granted greater leadership or committee roles if the bargain helps produce a new Democratic majority. (Democrats should do the same with state legislative maps, because the identical problem of over-packed majority-minority districts also exists in many states.)
In previous decades, African-American and Hispanic elites could hardly be blamed for doing what was necessary to break through the electoral glass ceiling. But the 2010 redistricting will be the fifth since the Supreme Court ruled malapportionment unconstitutional in the 1960s. Black and Hispanic legislators today are political enterprisers with their own personal constituencies, well-earned reputations, legislative accomplishments, and fundraising abilities. In addition to these advantages, do they really need districts so overwhelmingly packed with Democrats that they either run unopposed or against token opposition? What black and Hispanic Democrats in Congress need is more power. To obtain it, they should collude white Democrats in an effort to apply the Cummings-Wynn model wherever possible.
1Theoretically, it is also possible because populations of districts change during the course of a decade following reapportionment, as well as the complicating matter of at-large districts in the seven states which have only one House seat. But these differences are small; the main contributing factor is lower average turnout in Democratic districts.
Thomas F. Schaller is associate professor of political science at the University of Maryland, Baltimore County, and author of Whistling Past Dixie: How Democrats Can Win Without the South (Simon & Schuster, October 2006).