Which party best mirrors Texas in Austin?

When the 83rd session of the Texas legislature opens in Austin on Jan. 8, there will be a stark contrast between the ethnic/racial profiles of the Republican and Democratic delegations in the 150-member Texas House of Representatives. The 95-member GOP caucus is almost exclusively Anglo, with 88 Anglos (93 percent), three Hispanics (3 percent), 3 African Americans (3 percent), and 1 Asian American (1 percent).  On the other side of the aisle will be a much more diverse 55-member Democratic caucus, comprised of 30 Hispanics (55 percent), 15 African Americans (27 percent), eight Anglos (15 percent), and two Asian Americans (4 percent).

On the highly salient dimension of ethnicity/race, which party delegation best mirrors Texas depends on the population being reflected. The composition of the Democratic delegation in the House comes closest to reflecting the entire Texas population, while the Republican delegation comes closest to reflecting Texas registered voters and actual voters. The principal reason for these divergent findings is that while Anglos account for 45 percent of Texas residents, and Hispanics an increasingly proximate 38 percent , there still exists a substantial gap, albeit one that decreases each election cycle, between the proportion of Anglo and Hispanic registered voters (approximately 60 percent  to 25 percent ) and actual voters (approximately 65 percent  to 20 percent).

The past few weeks have featured extensive discussion of the limited support among Hispanic voters for the GOP in Texas (e.g., approximately 30 percent of Texas Hispanics voting for Mitt Romney). Often lost within this debate is a parallel discussion of the limited support among Anglo voters for Democratic candidates in the state (e.g., only about a quarter of Texas Anglos voting for Barack Obama). The analysis below highlights that while the House Democrats better reflect the population that represents the future of Texas, that is all Texas residents, House Republicans better reflect those Texans who today actually determine who wins elections in the state: registered voters and actual voters.

I use a common measure of disproportionality (the Loosemore and Hanby Index [LHI]) to compare the proportion of seats held by Anglos, Hispanics, African Americans, and Asian Americans in the Democratic and Republican House delegations with the share represented by these same groups within the state’s entire population (2010), population of registered voters (2010), and population of actual voters (average of 2008 and 2010). The calculation of the LHI involves taking the absolute value of the difference between each ethnic/racial group’s respective share of two distinct populations (e.g., legislative delegation and registered voters), summing these values, and then dividing this sum by two. A small LHI value indicates a high level of proportionality and a House delegation that closely mirrors the population in question  (be it all Texans, registered voters or actual voters).  A high LHI value indicates the opposite.

In the 2013 House, the Texas Democratic Party’s LHI for all Texas residents is 32, an identical value to that in 2011, but significantly higher (less representative) than in 2009 when it was 13. These values are all lower (more representative) than those of the GOP, which in 2013 had a LHI of 47, which itself was an improvement on 2009 when the LHI was 53. The Democratic delegation clearly better mirrors the state’s entire population than does the GOP delegation, although the gap between the two parties has narrowed substantially over the past four years, due primarily to the precipitous decline in the number of Anglo Democrats in the Capitol’s west wing.

In contrast to the above finding, the Texas GOP delegation more accurately reflects the state’s registered voters and actual voters than does the Democratic caucus. The 2013 Republican LHI for registered voters and actual voters are 34 and 29 respectively, while the comparable 2013 Democratic LHIs are 45 and 51. These numbers are comparable to those from 2011, but a sharp reversal from 2009 when the Democratic delegation’s LHIs for registered voters (25) and actual voters (29) were lower (more representative) than those of the GOP delegation (40 and 35 respectively). As was already noted, this reversal was not primarily the result of the GOP becoming more representative (though there was some modest improvement), but rather was principally due to the Democratic Party becoming less representative as a result of the losses suffered by the party’s principally rural Anglo representatives in the 2010 election.

Mark P. Jones is the Baker Institute’s fellow in political science as well as the Joseph D. Jamail Chair in Latin American Studies and chair of the Department of Political Science at Rice University.