Backers of the top two system argue that it will include more voters in securing who gets to to reach the general election. Right now, turnout in those primary elections can be extremely low, and independents are often kept out of voting at all. The result can be general election candidates who don't fully represent the electorate, allegedly contributing to polarizing partisans running legislatures.
But the top two system seeks to address this problem at the expense of party association rights (political parties typically lose the right to control who uses their name and how), voter choice in in higher turnout elections in November (if passed, Californians would only have at most two candidates to consider for each office in the final six months of their campaigns) and quirky distortions (if one party has four candidates run in the primary and the other party only two, the party with two might secure both general election positions). And, at the end of the day, representation is likely to look very similar. It's hard to imagine Democrats losing control of the California legislature very soon nor its caucus being much different in its composition as long as it;s elected in winner-take-all, single-member districts.
One of the longest experiences with the top two system at at state level is in Nebraska (for state legislative races, which are nonpartisan in its unicameral legislature) and Louisiana. We wrote about the Louisiana system in a 2004 analysis showcasing its troubling outcomes. (Note that Louisiana changed its top two system for congressional elections before the 2008 elections to a traditional primary, so the state's most surprising recent election result - the victory of Republican Joseph Cao in New Orleans last December -- was under a closed primary system, not the top two.)
As to Louisiana under the top two, its Members of Congress (and such governors as Mike Foster, who served two terms from 1995-2003) were not known as being particularly centrist. Consider the National Journal's ratings for Members of Congress in its 2004 almanac based on congressional votes in 2001-2002. These ratings of Members are based on what percentile of Congress their voting record on economic, social and foreign policy issues places them. For example, a rating of 4% liberal mean that only 4% of Members of Congress are more conservative than that Member. Here are the ratings for the state's US House Members:
- Vitter (R), district 1, who beat a more moderate Republican in a 1999 runoff in David Duke's congressional district (and in 2006 was elected U.S. Senator in the top two system)
* 2001 - Economic - 0% liberal - Social - 0% liberal - Foreign - 14% liberal
* 2002 - 0% in each area
- Jefferson (D), in district 2, long-time African American representative (who in 2008 lost due to corruption charges)
* 2001 - Economic - 68% liberal - Social - 79% liberal - Foreign - 83% liberal
* 2002 - Economic - 62% liberal - Social - 66% liberal - Foreign - 69% liberal
- Tauzin (R), in district 3, long-time representative who changed parties in the 1990s
* 2001 - Economic - 15% liberal - Social - 20% liberal - Foreign - 21% liberal
* 2002 - Economic - 0% liberal - Social - 0% liberal - Foreign - 15% liberal
- McCrery (R), in district 4, long-time representative
* 2001 - Economic - 13% liberal - Social - 38% liberal - Foreign - 4% liberal
* 2002 - Economic - 28% liberal - Social - 25% liberal - Foreign - 0% liberal
- Alexander (D), in district 5, and won open seat in 2002 so no voting record -- note that Alexander ultimately switched parties after running as a conservative in a very conservative district
- Baker (R), in district 6, long-time representative
* 2001 - Economic - 7% liberal - Social - 20% liberal - Foreign - 4% liberal
* 2002 - Economic - 19% liberal - Social - 0% liberal - Foreign - 0% liberal
- John (D), in district 7, moderate who is the only Member to "fit the profile" sought by advocates of the top two system. John was first elected in 1996 in a runoff against another Democrat (in an 8-person race where his leading Republican challenger finished 8 votes out of being 2nd)
* 2001 - Economic - 52% liberal - Social - 38% liberal - Foreign - 49% liberal
* 2002 - Economic - 53% liberal - Social - 48% liberal - Foreign - 52% liberal
The bottom line for FairVote is that if you really want to fully represent the spectrum, including centrist positions, you should do so with systems of proportional representation. That empowers voters wherever they live to elect candidates of choice. More on this point to follow!