by Karl Kurtz
Stateline's Josh Goodman has written a first-rate article on "The Disappearance of Multi-Member Districts." Focusing on Vermont's unusual six-member Senate district in Chittenden County (Burlington) and New Hampshire's large number of multi-member districts (MMDs), he does a nice job of outlining the reasons for creating MMDs (the representation of entire communities in the legislature) and why they have declined in number over time (court decisions finding that they cause racial discrimination).
I liked his paragraph about the need to consider the philosophy of representation in thinking about proposals to eliminate MMDs:
In Vermont, and in West Virginia as well, these proposals have prompted something rare: a debate about representation itself. Is it better that legislators have to appeal to a broad, diverse group of people, or should they represent a narrow, cohesive constituency? Are citizens better served by having a single representative who’s close to them or several different ones to whom they can voice concerns? While conventional redistricting politics is in play in these debates -- critics say, for example, that the multi-member Chittenden Senate seat has benefited Democrats -- the more serious philosophical questions are at least part of the conversation.
The article quotes my colleague Tim Storey about when the decline of MMDs began. His off-the-top-of-his-head statement that the steepest decline was in the 1980s is right on the mark, but it subsequently inspired the two of us to see if we could come up with the actual data on the number of chambers and states using MMDs. Here's what we found:
Does anybody have data from the 1960s?
The states that currently have MMDs are Vermont and West Virginia in both the house and the senate, Nevada in the Senate only, and Arizona, Idaho, Maryland, New Hampshire, New Jersey, North Dakota, South Dakota, and Washington in their houses of representatives only. Among the state houses that use MMDs, only Maryland, New Hampshire and West Virginia use it in some districts and not others with varying numbers of represenatives. In the others all representatives run from two-member districts, usually nested inside of state senate districts. (NCSL, Redistricting Law 2010, p. 137)
Peverill Squire and Gary Moncrief in State Legislatures Today: Politics Under the Domes (p. 27) summarize differences that political scientists have found in behavior between legislators in multi-member districts compared to those who represent single-member districts. Legislators in MMDs are more likely to:
- Regard themselves as "trustees" acting on the basis of what they think is best for their state and district, not necessarily the expressed views of their constituents
- Spend more time on constituent service
- Obtain more government funding for their districts.
Read below the jump for a 7/14/2011 update with more historical data on the number of chambers with MMDs.
These numbers are particularly interesting because they show within-decade changes as court decisions unfolded and states responded. Thank you, Pev!
Josh Goodman sent the following message:
"Thanks for the post! In my research I found an academic paper from 1966 that described which states used multi-member districts in 1960: http://www.jstor.org/pss/795047. Here’s what it said: “Up-to-date figures are difficult to obtain as systems change in efforts to conform to the reapportionment mandate. As of 1960, all of the state senators from Alaska and Arizona and some senators in sixteen other states were elected from multi-member districts. For the election of members of the lower house, Illinois, Maryland, and Washington used multi-member districts exclusively, and thirty-five states employed mixed systems of single and multi-member districts.” Of course, that doesn’t tell you how many multi-member seats there were in the 1960s after Reynolds v. Sims."
The numbers that Josh cites are actually the same as the 1950s data that I posted—the 35 states with mixed systems plus Illinois, Maryland and Washington equals 38. That makes sense because 1960 data are still part of the 1950s redistricting. Hence, I’m still looking for 1960s data.
Posted by: Karl Kurtz | July 14, 2011 at 10:01 AM