A "Competitiveness Index" for capturing competitiveness in state legislative elections
By Geoff Pallay and Leslie Graves
When we sat down in the summer of 2010 to figure out what factors we should use in constructing a Competitiveness Index that captures the extent of electoral competitiveness exhibited in state legislative elections, we realized that there are a number of different measurements that could be said to illuminate relative degrees of competitiveness. A case could be made for each of them as the most enlightening in constructing a robust multi-factorial scale that allows different states to be compared to each other, and the same state can be analyzed over different years.
Ultimately, we chose three key factors for our Competitiveness Index. We also collected data that we didn't use in constructing the Competitiveness Index. Much of this additional data is displayed throughout our "competitiveness analysis" pages available here on Ballotpedia, while some of it is on off-wiki spreadsheets. We invite any additional researchers who'd like to take a look at the information to e-mail Geoff Pallay for a copy.
In what follows, we discuss the three key competitiveness factors we chose in constructing the Competitiveness Index, as well as some that we didn't ultimately choose. We give our reasons for choosing the ones we did.
Three competitiveness factors used in Competitiveness Index
| Competitiveness Index analysts|
|Geoff Pallay Leslie Graves|
Factor 1: Is the incumbent running for re-election or not?
We chose to include this factor in our Competitiveness Index because when incumbents run, they enjoy significant advantages over their challenger. This is why, at all levels of politics, incumbents who are running for re-election defeat their challenger more than 90% of the time, even in years, such as 2010, when voters poll as having a high degree of anti-incumbent sentiment.
Everything else being equal, if you're looking at a state legislative chamber with 100 seats, and 100 of the incumbents are running for re-election, you have what we consider to be a "less competitive" election than if only 50 or 75 of the incumbents are running for re-election.
In the 2010 elections, our study showed that overall 86.6% of incumbents who were legally able to run in the election chose to run. Thus, only 13.4% of seats are considered open, with no incumbent running for re-election. We did a separate set of statistics for states where state legislators are term-limited, since this is going to have a significant impact on the relative number of open seats from state-to-state. However, in our overall chart that ranks all 46 state legislative chambers with 2010 elections, we straightforwardly include the term limited states. Although the number of open seats in those states is more due to term limits than to any other factor, it still results in a more highly competitive electoral environment.
Factor 2: If the incumbent does run for re-election, does he or she draw a primary challenge?
We've collected data that measures "If an incumbent is running for re-election, does he or she have a general election opponent" and "How many incumbents running for re-election had no opponent, either in the primary or in the general?", and we think these figures are extremely important to consider. (We are also somewhat shocked at the number of incumbents who had no electoral competition at all in 2010.)
But, for the purposes of the Competitiveness Index, we've elected to use an analysis of incumbents with primary challengers, versus incumbents without primary challengers. This factor shows interesting variance from state-to-state.
In our view, a state where relatively more incumbents face a primary challenge is clearly a more competitive electoral environment than those states with very few (and some, with no) primary challenges to incumbents.
Note that our 2010 study shows that in 2010 state legislative elections, only 22.7% of incumbents running for re-election actually faced a primary challenge. That means that 77.3% of incumbents running for re-election faced no primary opposition. Of the 1,130 incumbents who faced a primary challenge, only 88 lost their primary. This is 7.8% of the 1,130 incumbents who had primary opposition.
No major party competition in the general election
Factor 3: Are there two majority party candidates in the general election?
If a Republican or a Democratic candidate doesn't have a Democratic or Republican general election competitor, then 98% or more of the time, he or she is virtually guaranteed a win in November. A state where there are many legislative candidates without major party competition in the general election is clearly a state with a much less competitive electoral environment than states where most or all legislative candidates do have major party competition in the general.
Partisan breakdown of state senators
|Democratic state senators||1,021||51.8%|
|Republican state senators||895||45.4%|
|Nonpartisan state senators||49||2.5%|
|Independent state senators||3|
Partisan breakdown of state representatives
This chart of the party affiliations of state representatives as of September 2010 shows that only 21 are other than D or R.
|Democratic state representatives||3,015||55.9%|
|Republican state representatives||2,354||43.7%|
|Independent state representatives||21|
In most states, a large majority of the time, third party and independent candidates do not represent significant electoral competition to Republican and Democratic candidates.
For this factor, therefore, we simply counted how many Republicans have a Democratic opponent on November 2 and how many Democrats have a Republican challenger. We realize that there are some districts where a Republican or a Democrat does have serious competition from a third party or independent candidate but we believe that the number of these cases is so small as to not skew our overall results appreciably.
We intend to prepare an analysis of the competitiveness in state legislative elections in the years to come, and would very much appreciate commentary and feedback on our Competitiveness Index. Are there other factors that are more important that we didn't consider? Are we giving too much weight to the factors we did choose?
If you write about our Competitiveness Index on your website, we'll probably run across it, but just in case we don't, we hope you'll let us know.
Also, we decided to embark on this project as an outgrowth of our empirical observations about the 2010 legislative elections, which appear to us to have a surprisingly low degree of competitiveness, especially when contrasted with 2010's federal and state governor elections.
That said, we suspect that some scholars and researchers have done papers on this subject that we have yet to read and if we had read them, we might have made some different choices in constructing our Index. If you know of such papers, we hope you'll send us that information.
The data we used in assembling our Competitiveness Index and Analysis for 2010 was collected with the help of a wonderful team of summer interns:
...as well as with the devoted and careful work of our Ballotpedia staffers who focus primarily on state legislative content:
These individuals know more than they could have ever wanted to know about how often lists of candidates change and to them, we owe our undying appreciation.
- However, we'd be happy to take a look at individual districts where there are serious independent or third party candidates and to make note of those in our final report.