The Thicket at State Legislatures

Something Only a Legislative Junkie or a Redistricting Geek (They're not Always the Same) Could Love

Pennsylvania Senate Districts in 1790

Recently Pennsylvania Senate Majority Leader Dominc Pileggi and Senate Democratic Leader Jay Costa jointly announced the publication of historic Senate district maps dating back to 1790. Historic state House and congressional district maps are also available on the site. Here are some excerpts from their announcement:

"These maps are a window into our past," Senator Pileggi said. "Knowing this history can provide a deeper understanding of the issues confronted by previous General Assemblies through the decades. This is a tremendous resource, and I'm pleased it will be freely available for anyone interested in studying the history of the Senate."

"The Internet provides a platform where we can give the citizens of Pennsylvania greater access to state government," said Senator Costa. "This historical perspective will be an important resource. People will be able to see Pennsylvania's evolution from our very first legislative districts in 1790 to today."

The 18 newly-posted historic Senate maps feature current county lines, putting prior districts in a modern context. When Senate districts were first created in 1790, Pennsylvania had just 21 counties – fewer than a third of the 67 that exist today. Pennsylvania's final county, Lackawanna, was established in 1878.

Many changes in Senate districts have taken place over the 223 years that the chamber has existed. The first Senate included 12 districts and 18 senators – two districts were represented by three Senators each, and two other districts were represented by two Senators each. Such "multi-member" districts were used until 1874, when the state was apportioned into 50 Senate districts, the same number in use today.

 

Changes in Legislatures Using Multimember Districts after Redistricting

by Karl Kurtz

The Nevada Senate has eliminated the use of multimember districts (MMDs) in the new redistricting cycle, but the total number of MMDs has increased slightly, according to a recent compilation by my colleague, Morgan Cullen. Here is the lineup of legislatures with MMDs for the next decade:

Nevada's elimination of its two multimember districts in the Senate reduces the number of states with at least one MMD from 11 in the new redistricting cycle compared to 10 in the previous decade. But the total number of MMDs in the nation increased by five from 464 to 469. The increases were all in the New Hampshire (7) and Vermont (4) houses of representatives. This slight increase arrests (it would be hard to say reverses) a decades-long decline in the use of multimember districts in state legislatures.

Maryland, New Hampshire, South Dakota, Vermont and West Virginia are the only states that mix single member districts in some parts of the state with MMDs in other areas. The other states all have uniform two-member house districts, usually nested inside of a senate district. (In South Dakota's case, two out of 35 house seats have single member districts to comply with the Voting Rights Act and allow for representation of Native Americans; the remaining 33 seats are two-member districts.

The largest number of members in a house district is 11 in New Hampshire and in a senate district it's six in Vermont.

2012 Pre-election Turnover is Typical for a Redistricting Year

by Karl Kurtz

Ballotpedia, the useful wiki on state elections, recently published a paper, "Anti-Incumbency Voting Patterns in State Legislative Elections," that makes much of an apparently significant increase in primary election defeats of incumbent legislators in 2012 compared to 2010:

Anti-incumbent animosity is making itself felt in state legislative elections. Incumbent members of state legislatures in 2012 are being defeated in primary elections at a rate far higher than in 2010.... [I]n 2010, 8.38 percent of incumbents who faced a challenger in their primary were defeated, while the comparable figure for 2012 is 14.8 percent, for an increase of 76.7 percent.

This makes for a nice trend headline, and Ballotpedia has some great data on incumbent primary defeats to back it up. Trouble is , we know that 2012 is a redistricting year and that the years ending in "2" in every decade have significantly higher rates of turnover. Turnover is greater because more incumbents, faced with the challenge of contesting a newly-drawn district, are likely not to run again, and those who do are more susceptible to primary or general election defeat in their altered districts.

For this reason, I wished that Ballotpedia had compared 2012 primary election defeats to the last redistricting year, 2002, in addition to 2010. When I set out to do this myself, I discovered that we at NCSL do not have data on 2002 primary election defeats (and Ballotpedia wasn't around at the time). But we do have numbers for 2002 pre-election turnover--the combination of incumbent retirements and primary election defeats. By recalculating Ballotpedia's data for 2012, I came up with this table showing pre-election turnover for the 35 states that had held primaries as of Aug. 15, 2012:

8/22/12 correction: The following table has been changed from the original posting to correct data for North Dakota.

Sources: 2012 data from Ballotpedia.com recalculated by NCSL. 2002 data from NCSL.

 This analysis showing a marginal one percent nationwide difference in pre-election turnover between the two redistricting years results in a rather different headline: 2012 pre-election turnover in state legislative races, whether due to retirements or primary election defearts, appears to be typical for a redistricting year.

There are interesting state-to-state variations in these data. First of all, California, which has the highest rate of pre-election turnover, did not have a single incumbent primary election defeat--all of its turnover is due to term limits and retirerments. The effects of term limits (the "TL" column) on turnover are clearly evident in this table as eight of the top 10 states in 2012 have term limits.

In the term-limited states, the cycles of the limits come in different waves, which explains why Michigan and Missouri have much lower pre-election turnover in 2012 than in 2002. Perhaps this also explains how low the term-limited states of Ohio and Oklahoma are on the list in both 2012 and 2002, but I'm not entirely certain.

Kansas' high ranking on the list in 2012 is due to the extraordinary number of incumbent primary election defeats within the Republican party. But why are the rates of turnover in North Dakota more than twice as great in 2012 compared to 2002, given that only four incumbents lost in their recent primary? [8/22/12 correction] In Oregon the reverse is true--the turnover rate in 2002 was nearly twice as great as it is in 2012--but I don't know why.

Readers' contributions on these and other anomalies in the data are welcome.

On the (Non-)Effects of Redistricting on Electoral Competition

by Karl Kurtz

Six years ago in The Thicket, North Carolina's Gerry Cohen wrote a guest post, "Redistricting is Not the Cause of Non-Competitive Districts" based on academic research. I just now got around to reading another article on the same subject published early this year with an intriguing title: "The Gerrymanderers Are Coming! Legislative Redistricting Won’t Affect Competition or Polarization Much, No Matter Who Does It."

The abstract of the article by political scientists Seth E. Masket (University of Denver), Jonathan Winburn (University of Mississippi) and Gerald C. Wright (University of Indiana) provides a succinct summary:

Redistricting received substantial attention in the popular media in 2011, as states redrew state legislative and congressional district boundaries. Many reformers continue to argue for a de-politicization of the redistricting process, claiming that partisan redistricting is responsible for declining electoral competition and increasing legislative polarization. Our analysis of evidence from state legislatures during the last decade suggests that the effects of partisan redistricting on competition and polarization are small,
considerably more nuanced than reformers would suggest, and overwhelmed by other aspects of the political environment.

The article is relatively short and highly accessible to practitioners. Here's a sample figure showing the percentage of competitive state legislative seats (victory margin of 10 percent or less) after the 2000 redistricting cycle:

7,383

by Tim Storey

I would hazard a guess that only a very small number of The Thicket's most loyal readers could tell you, off the top of their heads, how many legislators serve in the 50 state legislatures. And most of those folks are sitting in offices here at NCSL. As of today, there are 7,382 state legislative seats plus another 195 in the District of Columbia and the territories and commonwealths. But that's about to change.

Every ten years following the redistricting cycle, a handful of states have changed the size of the legislature. In 1964 before the "reapportionment revolution" in legislatures, there were 7,838 legislative seats. In 1974, following the 1970's redistricting, there were 7,563 state legislators. In 1984, there were 7,438, and it dropped to 7,424 after the 90's redistricting.

It wasn't just redistricting that drove these numbers. In the 1960s and 70s legislative reformers argued that reducing the size of the legislature would make them more efficient and effective, and a number of states followed this advice.

7,382 has been the number of seats in legislatures for the past decade. But now, for the first time in almost 50 years, the number of legislators is about to increase...by one senator. As part of the new redistricting maps adopted by New York, the Empire State Senate will go from 62 to 63 seats. Some states have considered reducing the size of the legislature, but now, with the 2010 redistricting cycle mostly complete, New York appears to be the only state to alter the size of either chamber.

So, with the addition of one new desk on the floor of the New York Senate following this fall's election, there will be 7,383 total legislative seats in the United States for the next ten years.

Those 7,383 legislators will represent nearly 314 million Americans. That's an average of roughly 42,500 constituents per state legislator in 2013. Compare that to 100 years ago when the population was less than one-third what it is today and there were about the same number of legislators (in only 48 states). In 1912, there were 7,257 legislators representing an average of 12,700 constituents. 

The table below shows changes in size of the total of all legislative seats from decade to decade back to the 1940s.

A Majority of State Legislators May Have Two Years or Less of Experience in 2013

by Karl Kurtz and Tim Storey

The combination of unusually high turnover in the membership of state legislatures after the 2010 election and anticipated high rates of churn after the redistricting election of 2012 makes it likely that approximately half of all state legislators will have served for two years or less at the start of 2013 legislaitve sessions. The 2010-12 cycle almost certainly will result in the highest rate of turnover in state legislatures over two elections in the last 50 years.

How do we arrive at this conclusion? The 2010 nationwide surge of Republican gains in state legislatures resulted in a high water mark of 29 percent for membership turnover in a non-redistricting year. For the upcoming 2012 election, the average rate of turnover in legislatures in years ending in "2" in the last three redistricting cycles is 30 percent. We have no reason to believe that 2012 will be any different.

Combining these two election results, it seems likely that about half of all legislators will be freshman or sophomore members in their chambers. Unfortunately, though, we can't simply add together the numbers from the two elections to produce a more definite number and avoid weasel words like "may" or "seems likely." There are four reasons for this:

1. The 2012 election could turn out to be different from previous post-redistricting elections. Predicting any kind of election results is notoriously difficult.
2. Many of the seats that will turn over in 2012 also had new members after the 2010 election. The result in each of these districts will be a legislator elected with no previous experience, but if we add 2010 and 2012 turnover together, we would be counting that seat twice in our tally of inexperienced members in 2013.
3. Some new members elected to one chamber in a state will have had experience serving in the other chamber, so we must be careful not to exaggerate the inexperience levels.  For example, 29 of 38 members (76 percent) of the Michigan Senate were new to that body after the 2010 election. But 14 of them had previous experience in the House, so the real rate of legislative inexperience in the Senate was 39 percent (15 of 38 members). 
4. In a relatively small number of cases, legislators with previous experience who have been out of the legislature may return to run again. This is especially likely in the nine states with term limits based on consecutive service (not lifetime limits). For example, despite eight year (consecutive) term limits for the Ohio House, Speaker William Batchelder is now in his 37th year of legislative service because he sat out for a while and then returned to the General Assembly. Our turnover data would have counted him as a new member in the year that he returned.

Despite these caveats, we are confident in our conclusion that very close to half or more of all state legislators will have less than two years of experience after the 2012 election. Of course, not every state will have high levels of inexperience, but it is a warning to leaders and legislative staff (and NCSL and other national organizations) that there is a need to expand training and professional development opportunities for the influx of freshman and sophomore legislators.

Coincidentally, over at Governing magazine Lou Jacobson posted a column on the same subject today. It's a good piece that goes into more depth about anticipated turnover in specific states and the consequences of inexperience. We take a somewhat different view of the causes of this unusual two-election turnover cycle, though. We suggest that it is more due to the Republican sweep of 2010 and redistricting in 2012 than it is term limits and redistricting. Term limits have of course increased turnover, but the waves of turnover come in different years in different states and cannot by itself account for the anticipated 2010-12 turnover phenomenon.

Florida Explores New Frontier in Redistricting

by Morgan Cullen

This week Florida lawmakers reconvened in Tallahassee for a special legislative session on redistricting. Last Friday, the Florida Supreme Court upheld the Legislature's House plan in its entirety but struck down the Senate redistricting plan. For the first time, the Court was required to interpret a new constitutional amendment that prohibits state legislative redistricting plans from "favoring or disfavoring an incumbent or political party."

In November of 2010, Florida voters passed the "Fair Districts Amendment" to their constitution prohibiting political and incumbent gerrymandering. The amendment also prohibits drawing districts that discriminate against minority voters by "denying them an opportunity to participate in the political process or diminishing their ability to elect a candidate of their choice." Additionally, it requires districts to be contiguous, equal in population, compact and to maintain "political and geographical boundaries where feasible."

Florida is not the only state with standards that limit partisan or incumbent favoritism in redistricting. Currently 12 states have some form of prohibition against plans that favor incumbent lawmakers or a particular political party. But these limits vary from constitutional standards in some states to committee guidelines in others. The states with the strongest restrictions contained in state constitutions are Arizona, California and Florida. Arizona's Constitution uses a positive, rather than negative, approach: it requires all districts to be "competitive" to the "extent practicable." Federal law does not place limits on drawing plans that advantage a particular party or individual.

Legislative Scholar is State's Top Politico

by Karl Kurtz

Quick. Name an academic political scientist who has spent a career studying state legislatures but has also recently been named the most powerful person in his state's politics during the last year.

Does that seem like an unlikely combination? Try Rutgers University Prof. Alan Rosenthal who, in addition to writing numerous influential books on state legislatures, has served on two different occasions as the 11th, nonpartisan, tie-breaking member of New Jersey's state legislative redistricting commission. For his decision last year to back a plan favored by Democrats, in the face of fierce pressure from Gov. Chris Christie, Politicker NJ ranked Rosenthal #1 among New Jersey's 100 most powerful politicos, excluding elected officials, in 2011.

PolitickerNJ's citation said of Rosenthal: "This academic succeeded this year where every Democratic politico in the state failed; he checked the power of Gov. Chris Christie." Rosenthal's rationale was that he chose the plan that was more likely to allow either side to win control of the legislature and was "less disruptive" to incumbents.

When asked about the award, Rosenthal laughed and said, "It's pretty amazing because there were a bunch of people on the list below me who are genuinely powerful."

Photo credit: PolitickerNJ

Federal Court Rejects Texas Legislature's Claim of Legislative Privilege

by Karl Kurtz

Earlier this week a federal district court judge in Washington, D.C. ordered the Texas Legislature to hand over documents that it had withheld in a dispute over redistricting maps. The Justice Department and plaintiffs sought access to the documents prepared by legislators and legislative staff. The Legislature claimed that the documents were protected by legislative privilege and an attorney-client relationship between the staff of the Texas Legislative Council and the members of the Legislature.

Legislative privilege is the doctrine found in the federal Constitution and 43 state constitutions that provides legal protection to legislators in the conduct of legislative business. The intent is to shield them from threats of judicial or executive intervention in their work.

The court rejected the Texas claim of legislative privilege, saying that the Legislature had failed to cite any statute or case law that would justify the privilege. Moreover, the court said that while it recognizes the confidentiality between staff and legislators provided for in the Texas Government Code, that confidentiality does not constitute an attorney-client relationship. Finally, the court ruled that communications between legislative staff and members about the case had to be disclosed because the legislators involved did not claim privilege (the court did not say how it would have ruled if they had claimed privilege).

The Texas Redistricting blog has a more extensive summary of the case and links to the court's decision.

To my inexpert and non-legal eyes, this case seems to hinge mainly on the language of Texas statutes and case law, so it may not have major implications for questions of legislative privilege and attorney-client relationships in other states. However, at the very least it seems to be a case of a federal court joining with state courts in a tendency to side with government openness and transparency against the competing value of the right of legislatures to maintain independent decision-making processes. Ohio State law professor Steven F. Huefner reports in a 2003 law review article, "The Neglected Value of the Legislative Privilege in State Legislatures":

While some of these states' [legislative privilege] provisions have never been applied, recent judicial interpretations in other states have departed from settled federal interpretations of the legislative privilege, failing to apply it broadly to protect the legislative process and instead unduly favoring ideals of open government.

He goes on to conclude:

...To protect representative democracy, the legislative privilege merits a more robust application at the state level than some state courts have been willing to give it.

It now appears that Prof. Huefner can add the federal courts to his plea.

In addition to the Huefner article, my colleague, Kae Warnock has compiled some useful resources on the subject of legislative privilege, legislative attorney-client relationships and confidentiality:

• "Legislative Immunity in Minnesota" by former Senate Counsel Peter Wattson includes extensive citations of federal and state case law (not just Minnesota).
• "Representing Representatives: Ethical Considerations for the Legislature's Attorneys," a 2003 law review article by former Wisconsin staffer Rob Marchant
• Kae's own informal compilation of confidentiality policies and attorney-client privilege statutes and case law.

The Texas Redistricting Case Explained

by Karl Kurtz

We haven't written about the U.S. Supreme Court accepting the Texas redistricting case in large part because it has been in the national news. But I've read half a dozen stories about the challenge that were either inaccurate or so murky that it was hard to make sense of the case--which is complex to begin with. Yesterday, though, the Washington Post's Chris Cillizza wrote "Texas redistricting case: Five things you need to know." It's a tidy summary of an untidy case and recommended reading. It includes this speculation on two of the more extreme possible outcomes of a Supreme Court ruling:

The real question, then, is what kind of instruction the Supreme Court offers. Those watching the process suggest two potentially far-reaching outcomes (among many other possibilities, which are too numerous to name).

First, the Supreme Court could tell the three-judge panel in San Antonio that it needs to draw something closer to the map drawn by the state legislature. Or second — and this is the Nuclear Option — it could say that the state should simply use the map drawn by the state legislature.

The former option would set a new standard when it comes to court-drawn maps. Currently, court-drawn maps are drawn with deference to the last constitutionally-approved map available (i.e. the existing map). Changing the standard would give state legislatures greater power over the final product, even in the event that their maps are invalidated.

The latter would be the big one, as some suggest it would essentially invalidate the section of the Voting Rights Act — Section 5 — that requires states like Texas to get pre-clearance. Essentially, the maps would no longer have to be pre-cleared. This would allow these states greater freedom to draw their maps and increase the burden on minority groups and others who may fight the maps in court. These groups would no longer have an avenue to stop the maps before they are enacted.

This is the first redistricting case that SCOTUS has accepted in the 2010 redistricting cycle and the first case since the high Court ruled in Bartlett vs. Strickland (North Carolina) in 2009.