Top Banner
1 Redistricting and Polarization Micah Altman and Michael McDonald * Corrected Manuscript Forthcoming in American Gridlock: The Sources, Character, and Impact of Political Polarization, 2015, James A. Thurber, Antoine Yoshinaka (Eds), Cambridge University Press. We review how the ideological polarization of members of the House of Representatives (elite polarization) is affected by: sorting of parties’ incumbents into more ideologically compatible districts, replacement of incumbents by more ideologically extreme successors, * Authors are listed in alphabetical order. We describe contributions to the chapter using a standard taxonomy (Allen et al. 2014). Micah Altman and Michael McDonald were the lead authors, taking equal responsibility for revisions. Michael McDonald authored the first draft of the manuscript and was primarily responsible for the statistical analysis. Both contributed to the conception of the report (including core ideas and statement of research questions), to the methodology, to the project administration, to the data collection, and to the writing through critical review and commentary.
38

Redistricting and Polarization

Jan 03, 2017

Download

Documents

tiet nhan
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Redistricting and Polarization

1

Redistricting and Polarization

MicahAltmanandMichaelMcDonald*

CorrectedManuscriptForthcomingin

AmericanGridlock:TheSources,Character,andImpactofPoliticalPolarization,

2015,JamesA.Thurber,AntoineYoshinaka(Eds),CambridgeUniversityPress.

• WereviewhowtheideologicalpolarizationofmembersoftheHouseof

Representatives(elitepolarization)isaffectedby:

• sortingofparties’incumbentsintomoreideologically

compatibledistricts,

• replacementofincumbentsbymoreideologicallyextreme

successors,

*Authorsarelistedinalphabeticalorder.Wedescribecontributionstothe

chapterusingastandardtaxonomy(Allenetal.2014).MicahAltmanandMichaelMcDonaldweretheleadauthors,takingequalresponsibilityforrevisions.MichaelMcDonaldauthoredthefirstdraftofthemanuscriptandwasprimarilyresponsibleforthestatisticalanalysis.Bothcontributedtotheconceptionofthereport(includingcoreideasandstatementofresearchquestions),tothemethodology,totheprojectadministration,tothedatacollection,andtothewritingthroughcriticalreviewandcommentary.

Page 2: Redistricting and Polarization

2

• thedrawingofmoreideologicallyextremedistricts.

• Weshowtherearefewercompetitivecongressionaldistricts–havinga

nearbalanceofDemocratsandRepublicans–followingredistricting.

• Weshowthatmorecompetitivedistrictscanbedrawnwithout

sacrificingothervalues,suchascompactnessorminorityrepresentation.

• Wediscusstheprospectsforredistrictingreform.

IncreasingideologicalpolarizationofAmericanpoliticalelites(“elitepolarization”)

hascoincidedwithamarkedincreaseofpolicygridlockwithinthenational

government.Concernswiththegovernment’sabilitytoaddressmajorpolicyissues,

eventhosewithbroadpublicsupport,haveledmanytosearchforcausesforand

solutionstoagovernmentsoparalyzedthatitcannotsatisfythebasicdemocratic

valueofexecutingthewillofitspeople.

Amongthefrequentlycitedcausesforelitepolarizationisredistricting,the

processofperiodicallydrawingdistrictboundariestoostensiblyalignthemwith

communitiesofinterest,representationalcriteria,andneutraladministrativegoals,

suchasequalizingpopulationsfollowinganewdecennialcensus.Statesare

responsiblefordrawingHouseofRepresentativesdistrictboundarylines;inmost

states,politiciansareinchargeoftheprocess,creatinganobviousconflictof

interestbecauseredistrictingaffectstheirchancesofreelection.

Redistricting’spotentialcontributionstoelitepolarizationemergefromthe

motivesofindividualpoliticiansandpoliticalparties.Districtboundarylinesmaybe

manipulatedtoaffectelectionoutcomesbyshoringupanotherwisecompetitive

Page 3: Redistricting and Polarization

3

district(adistrictwithanearbalanceofDemocraticandRepublicanvoters)by

addingsupportersofthepartyonewishestoadvantageorsubtractingthe

opposition.Inredistricting,thereareopportunitiesforbipartisanshipthatelude

nationalpolicymaking:oneincumbent’strashisanother’streasure,whereby

incumbentsofdifferentpartiesswapconstituentsunfavorabletotheirparty,but

supporttheother.Iflegislatorsreflectthewilloftheirconstituents,elite

polarizationincreaseswhendistrictsaremademoreideologicallyhomogeneousas

reelection-seekingincumbentshavelessreasontomoderatetheirpositionstowin

overindependentsortheotherparty’ssupporters.

Reformershaveamendedtheredistrictingprocessinahandfulofstatesto

imposelimitations,includingmostrecentlyinNewYorkthrougha2014

constitutionalreferendumapprovedbyvoters.Amongthetechnicalinnovations

thatquietlyemergedduringthelastroundofredistrictingispublicaccesstoweb-

deployedredistrictingsoftwareanddata,whichenabledgreaterpublicparticipation

inthedrawingofalternativeredistrictingplans(AltmanandMcDonald2014b).The

publicapproachesredistrictinginafundamentallydifferentmannerthan

politicians,particularlywithrespecttopoliticalgoalssuchaspartisanfairnessand

districtcompetition(AltmanandMcDonald2013,2014a,2014c).Thesepublicplans

therebyprovideabenchmarkcomparisontoplansfabricatedbypoliticiansto

concretelyassesshowdistrictsmaybemademorecompetitivethroughreform

effortsandtowhatdegreethismayamelioratepolarization.

Page 4: Redistricting and Polarization

4

The Causal Relationship between Redistricting and Polarization

Figure 1 Causal Connections Affecting Elite Polarization

Muchhasbeenwrittenaboutthecausesandconsequencesofpoliticalpolarization,

includingbythecontributingauthorstothisvolume.Ourpurposeisnottorehash

theextensivepolarizationliterature.However,wedowishtoprovidecontextfor

redistricting’srole.InFigure1wesketchhowredistrictingfitswithintheoretical

frameworksexplainingelitepolarizationoftheHouseofRepresentatives.The

linkagebetweenredistrictingandelitepolarizationismediatedthroughdistrict

partisanship,asvisualizedinFigure1.Thecausalchainfromredistrictingtoelite

polarizationhastwonecessaryconditions.First,theideologicalcharacterof

districts’constituenciesmustberelatedtotheideologicalcharacteroftheirelected

representatives.Second,redistrictingmustbeabletoaffectdistricts’ideologies.We

beginourdiscussionofthefirstordercausalconnectionswithcausesthatdirectly

Page 5: Redistricting and Polarization

5

affectelitepolarization,andthenwithsecondordercausalconnectionsaffecting

thesefirstordercausalconnections.

First Order Connection: District Partisanship → Elite

Polarization

Withrespecttothefirstcondition,therearenoreliablenationalsurveymeasures

fortheideologyofdistricts’constituenciessincesamplesizeswithindistrictsare

typicallytoosmalltodevelopreliablemeasures.Scholarsfrequentlymeasure

districts’ideologiesusingdistricts’partisanship,constructedfromelectionresults

forstatewideoffices(McDonald2014).Presidentialelectionresultsareanalyzedfor

nationalstudiessincethisisthesoleofficeelectednationally,thusprovidinga

comparablemetricacrossallcongressionaldistricts.Wefollowthescholarlynorm

equatingdistrictideologywithdistrictpartisanship.

Themedianvotertheoremprovidesatheoreticalbasistoexpectthatdistrict

partisanshipisafactorinrepresentatives’ideologies.Downs(1957)formallyshows

howinplurality-wintwo-candidateelections,candidateswhowishtowinoffice

will,inequilibrium,positiontheirideologysuchthatitisthesameasthedistrict’s

medianvoter’sideology,inanelectoratewhereallvotersaredistributedalonga

one-dimensionalleft-rightideologicalcontinuum.(Thisis,ofcourse,acaricatureof

Downs’snuancedtheory;wediscusscomplicationsandothermotives

subsequently.)

Page 6: Redistricting and Polarization

6

Thelogicissimpleand,intheabstract,plausible.Consideracandidatewhose

primarymotivationistogetelected.Acandidatewinsofficebyreceiving50percent

plusoneofthevotes.Ifvoterspreferthecandidatewiththeclosestideology,then

withoutknowingwhattheothercandidatewilldo,thefirstcandidateisalwaysbest

offpositioninghisorherideologyatthemedianvoter.Therearetwocasesto

consider.Caseone:ifthesecondcandidatepositionsatanypointotherthanthe

median,thefirstcandidateonthemedianissupportedbyallvoterstotheleftor

right(thosewithanideologyintheoppositedirectionfromthemedianasthe

secondcandidate’sposition),aswellashalfofthevotersbetweenthemedianand

theideologyofthesecondcandidate.Thefirstcandidatewinsandthesecondloses.

Casetwo:ifthesecondcandidatealsopositionsatthemedianvoter,thenvotersare

indifferentbetweenthetwocandidatesandtheyrandomlychoosebetweenthetwo

candidates.Afirstcandidatewhochoosesapositionotherthanthemedianis

gamblingthattheothercandidatewillpositionhis-orherselfevenfurtherfromthe

median.Whytakethisriskwhenthereisaclearpathwaytovictorybypositioning

atthemedian?

Theredistrictingapplicationisstraightforward.Theideologyofthemedian

voterofadistrictisrelatedtotheproportionofvotersontheleftandright(i.e.,

DemocratsandRepublicans)whoareassignedtoadistrict.Ifmorevotersonthe

leftorrightareaddedtoadistrict,theideologyofthemedianvoterofthedistrict

willbeavoterfurthertotheleftorright,respectively.Thoseinchargeof

redistrictingthushavetheabilitytoaffectnotonlywhichpartyisfavoredtowina

Page 7: Redistricting and Polarization

7

district,but,bymanipulatingtheproportionofvotersinadistrictontheleftor

right,theideologicalcharacterofthewinningcandidateaswell.

Toempiricallydemonstratetherelationshipbetweendistrictpartisanship

andelitepolarization,whichwewillrefertoagainwhenwedescribeothercausal

pathways,weplotameasureofdistrictpartisanshipagainstameasureofmembers’

ideologyinFigure2.OurmeasureofdistrictpartisanshipistheMcCainshareofthe

ObamaandMcCain2008presidentialvote,orwhatiscommonlycalledthetwo-

partyvote.AhigherpercentagethusdescribesamoreRepublicandistrict.Wemight

reasonablyinferthatdistrictpartisanshipproxiesconstituents’ideology,witha

moreRepublicandistrictsignalingamoreconservativeconstituency.Ourmeasure

ofmembers’ideologyisPooleandRosenthal’s(2000)firstDW-NOMINATE

dimension,whichplacesmembers’rollcallvotesonaleft–rightcontinuum,with

highervaluesrelatedtoamoreconservativerollcallvotehistory.NOMINATE

scoresareendogenoustoCongress,meaningthattheyarebasedonobservableroll

callvotesthataretheendproductofthelegislativeprocess,andmaynotbetrue

measuresofmembers’ideology.Ifonecaresaboutelitepolarizationwithin

Congress,theyserveasastaticsnapshotofthecurrentCongress.Wefurther

identifyDemocratswithaDandRepublicanswithanR,anddrawasimplelinear

regressionlineforeachpartisantype.

TherearethreeimportantfeaturesofFigure2:(1)thecorrelationbetween

districtpartisanshipandmembers’ideologies,(2)partydivergenceofmembers

fromdifferentpartiesrepresentingdistrictswithsimilarpartisanship,and(3)

Page 8: Redistricting and Polarization

8

candidatedivergenceamongmembersofthesamepartyrepresentingdistrictswith

similarpartisanship.Wefirstdiscussthefirstfeatureandreturntothetwoothers.

Figure 2 District Partisanship and Representatives’ Ideology

Centraltotheargumentthatredistrictingcanaffectpolarization,and

consistentwiththemedianvotertheorem(Downs1957),thereisastronglinear

relationshipbetweendistricts’partisanshipandmembers’ideologicalvoting

patterns,withmembersfrommoreRepublicandistrictsvotinginamore

Page 9: Redistricting and Polarization

9

conservativemanner.Figure1isnotcontroversial.Thispatternofcandidate

convergencetothedistrictideologyispresentedintextbooks(Abramsonetal.

2014)andconfirmedbyananalysisofallcandidates’ideologies–incumbents,

challengers,andopenseatcandidates–derivedfrommembers’ideologicalvoting

scores(RyanandLyons2014)andcandidatesurveys(EriksonandWright1980;

Ansolabehere,Snyder,andStewart2001;Stonecashetal.2003;).Thisconvergence

isevidentinotherlegislativeactivities,suchasbillsponsorshipandrollcallvoting

withinspecificissuedomainsthatmaybesensitivetoconstituencydemographics

(Hayes,Hibbing,andSulkin2010).Ansolabehere,Snyder,andStewart(2001)find

candidatesmorecloselyhewtodistrictpartisanshipinthemostcompetitive

districtsnear50percent(thisheteroskedasticdynamicismostapparentamong

RepublicanincumbentsinFigure1),althoughtheyalsofindthatchallengersto

incumbentslesscloselyconvergetodistrictpartisanshipthanincumbents.

First Order Connection: Party Divergence → Elite

Polarization

AsecondfeaturethatisillustratedbyFigure2isasizableideologicalgapbetween

thetwopoliticalparties.Anincreasingideologicalgapbetweenthepoliticalparties

iswelldocumentedandhasgrownsomuchthatthereiscurrentlynooverlap

betweenthemostconservativeDemocratandmostliberalRepublican(McCarty,

Page 10: Redistricting and Polarization

10

Poole,andRosenthal2009;seealsoChapter16,thisvolume).Contributingtothis

elitepolarizationisadeclineinthenumberofincumbentswhorepresentdistricts

withavotercomposition–measuredinpresidentialvote–favoringtheopposing

party,aphenomenonthatcontinuesfollowingthe2014election.Thisdecreaseisa

consequenceofthesortingbroughtaboutthroughthesectionalrealignmentofthe

SouthandNortheast.Beforetherecentriseinelitepolarization,conservative

SouthernDemocratsandliberalNorthernRepublicansfilledtheideologicalcenter

betweenthemoreextremeelementsoftheirrespectivepartycaucuses.Inrecent

decades,SouthernDemocratseitherformallyaffiliatedthemselveswiththe

RepublicanPartyoraRepublicanclaimedtheirseatfollowingaretirementor

electoraldefeat(settingasideSouthernDemocratsrepresentingminority

communities).WhileregionalrealignmentwasmostpronouncedintheSouth,a

mirroreddynamicoccurredamongliberalRepublicansintheNorth.Redistricting

mayalsoaffectsorting,whenincumbentsmismatchedwithanunfriendly

constituencyaredrawninanew,ideologicallycompatibledistrict(Jacobson2003;

McCartyetal.2009);wediscussthisingreaterdetaillaterinthechapter.Parties

havethusbecomemoreideologicallyconsistent(FiorinaandPope2010;Laymanet

al.2010),whichresultsinafurtherfeedbackmechanismwherebymembersofmore

ideologicallyhomogeneouspartycaucusesprovidetheirleadersadditionalpowers

toenforcepartydiscipline,whatisknownastheconditionalpartygovernment

model(Rohde1991).

AcaricatureofDowns(1957)portraysthemedianvotertheoremas

predictingcandidates’fullconvergencetothemedian,framingdivergenceasa

Page 11: Redistricting and Polarization

11

“failure”ofthemodel(Ansolabehere,Snyder,andStewart2001:153).Downsdid

notpredictfullconvergence,ashecouldobservetheobviousfact–evenin1957–

thatinAmerica’stwo-partysystemtheparties’candidateswerenotideological

clones.Downspositedthatpartieswouldnotadoptthemedianpositionwithina

districtinordertodistinguishthemselvesfromoneanotherandtherebypresent

voterswithtangiblechoices.Butthereareothercompellingreasonsfordivergence.

AsevidentfromthedistributionofdistrictsacrosstheX-axisinFigure2,districtsdo

nothavethesamepartisanship,socandidateswithinpartycoalitionswouldbehard

pressedtosimultaneouslyadoptindividualideologiesconsistentwithan

overarchingnationalpartyideology(seecontributionsinGrofman,Blais,and

Bowler2009).Theremaybeotherreasonsforthisdynamicaswell,suchas

candidateshavingintrinsicpolicypreferencesoftheirown,beingreputationally

boundtotheirpreviouspolicypositions,beingresponsivetobothmore

ideologicallyextremeprimaryandmorecentristgeneralelectionelectorateswithin

thesamedistrict(Brady,Han,andPope2007),orthatentrycostsdeterpolitically

moderatecitizenswhomaywishtorunforoffice(GrosserandPalfrey2014).

First Order Connection: Candidate Divergence → Elite

Polarization

AthirdfeatureillustratedbyFigure2isthatwithinpartiesthereissizablevariation

ofideologyevenamongmembersindistrictswithcomparabledistrictpartisanship.

ThisismoreevidentintheSenate,wheresenatorswhoareofthesamepartyand

Page 12: Redistricting and Polarization

12

whorepresentthesamestatehavedifferentideologies(PooleandRosenthal2000).

Itisalsoevidentinhowmembersfromthesamepartyreplacingaretiring

incumbentdonotadoptthesameideology(PooleandRomer1993;Stonecash,

Brewer,andMariani2003;Theriault2006).Indeed,thesereplacementshave

tendedtoadoptamoreextremepositionthantheirpredecessor,thereby

contributingtopolarizationwithmoderatesbeingreplacedbyextremists(Bafumi

andHerron2010).Incumbentsmayalsoadapttochangingpoliticalenvironments

bymodifyingtheirideologicalvotingoverthecourseoftheircareers(Theriault

2006),withmembersmovingtowardtheirdistrictpartisanship(Stratmann2000).

Itisforthislatterreasonthatwedrawacausalarrowfromcandidatedivergenceto

elitepolarization,althoughFigure2snapshotsuggeststhereismuchideological

variationamongmembersrepresentingdistrictswithsimilarpartisancomposition,

whichisnotnecessarilycontributingtoideologicaldivergenceofparties.

Second Order Connection: Redistricting → District

Partisanship

Thesecondconditionnecessaryforredistrictingtoaffectelitepolarizationisthat

redistrictingaffectsdistrictpartisanship.Itmayseemself-evidentfromobserving

theintensepoliticsthatsurroundredistricting,andtherelatedproblemof

gerrymandering,thatacausalconnectionbetweenthesetwoexists.Indeed,many

scholarsfindredistrictingaffectspartisandivisionofpoweramonglegislativeseats

(e.g.,Tufte1973;Cain1985;GelmanandKing1994;Hirsch2003),whileothers

Page 13: Redistricting and Polarization

13

haveproposedevaluationmetricsonhowmuchpartisangerrymanderingmaybe

constitutionallypermissible(GrofmanandKing2007).However,thissecondorder

effectischallengedbycompetingexplanationsforchangingdistrictpartisanship,

particularlythenumberofcompetitivedistrictsthat,throughtheconnectionof

districtpartisanshipandmembers’ideologies,driveelitepolarization.

Abramowitz,Alexander,andGunning(2006)andMcDonald(2006a)differin

theirconclusionsaboutwhetherredistrictingresultsinfewercompetitivedistricts.

Themeasurementofcompetitivecongressionaldistrictsusingpresidentialelections

liesattheheartoftheirdivergentfindings.Analyzingnormalizedtwo-party

presidentialvote,1Abramowitzetal.(2006)findanincreaseincompetitivedistricts

duringthepost-1990censusbetweenthe1988Bushvs.Dukakiscontestandthe

1992three-wayBushvs.Clintonvs.Perotcontest(excludingPerot’svotessincehe

wasnotamajorpartycandidate).However,inanalyzingtheBushandDukakisvote

inthepre-andpost-redistricteddistricts,McDonald(2006a)findsadecreaseinthe

numberofcompetitivedistricts.2

WeupdateMcDonald’s(2006a)analysisofthetotalnumberofcompetitive

districtsbeforeandafterredistrictingforthemostrecentroundofredistrictingin

Figure3.Thetablereportsthenumberofdistrictswithintwocompetitiveranges,

1“Normalizedtwo-partyvote”isthevoteshareforamajorpartycandidate

expressedasashareofthetwomajorparties’candidates,excludingminorpartycandidates.Thisvoteshareisthennormalized,orsimulatingahypothetical50/50election,bysubtractingthedifferencebetweentheoverallnationaltwo-partyvoteshareand50percentfromthetwo-partyvotewithinaspecificdistrict.

2Forsimilarfindingsinthedeclineincompetitionfollowingaredistricting,seeSwain,Borrelli,andReed(1998)andMcCartyetal.(2009:673).

Page 14: Redistricting and Polarization

14

45–55percentand48–52percent(thelatterareincludedintheformer).Thevote

betweenthetwomajorpartycandidates(commonlyreferredtobyscholarsasthe

“two-partyvote”)isdrawnfromvariousissuesoftheAlmanacofAmericanPolitics.

Thepresidentialvotesharesarenormalizedbysubtractingthedifferencebetween

theleadingcandidate’svoteshareand50percenttosimulateahypothetical50/50

election.Tocontrolforpotentialconfoundingcampaigneffects,statisticsare

calculatedforthesamepresidentialelection,theonemostrecentlyoccurredprior

toagivenredistricting.

Figure 3 Competitive Districts Before (Red) and After (Blue) Redistricting, 1980–2012

Page 15: Redistricting and Polarization

15

Take,forexample,the2008presidentialelection,whoseresultsarereported

forthe2010and2012statistics.ThestatisticsinFigure3showthatforthemost

recentredistricting,therewasaslightdecreaseoffourdistrictswithinthewider

competitivenessrangeandadecreaseoftwelvedistrictsinthenarrower

Page 16: Redistricting and Polarization

16

competitivenessrange.Despitethetraditionalrelianceona45–55percenttodefine

competitionelections(Mayhew1974),McDonald(2006a)findsfromastatistical

analysiscorrelatingpresidentialvoteandcandidatevotesharesthatthenarrower

rangeismoreappropriatetomeasurecompetitivedistricts.Bothnumbersarenear

thelowerboundevidentinthelasttworedistrictingcycles.Inthelargerpicture,

exceptforthe1980cycle,therewasadecreaseinthenumberofcompetitive

districtsinthenarrowerrangeinthe1990,2000,and2010redistrictingcycles,and

forthewiderrangetherewasalsoandecreasein1990,2000,and2010butan

increasein1980.UsingAbramowitzetal.’s(2006)preferredmeasure,the2012

normalizedtwo-partyvote,comparedtothe2008presidentialvote,only

strengthensthesefindings.3Thepreponderanceoftheevidenceshowsredistricting

resultsinfewercompetitivedistrictsinthepastthreedecades.

3Anissueforthepost-redistrictingstatisticsisthatsomeprecinctsaresplit

bynewdistricts,andthevoteswithinthesedistrictsmustbeapportionedtothenewdistricts(McDonald2014).Usingthe2012presidentialvotesharesforthe2012statistics,thenumberofdistrictsis84withina45–55percentrangeand29withina48–52percentrange.Thisalternativemeasurementprovidesstrongerevidenceforredistrictingtoresultinadecreaseinthenumberofcompetitivedistricts.WhereasthestatisticsreportedinFigure3showamodestdecreaseoffourinthenumberofcompetitivedistrictsinthewiderrangebetween2010and2012,thealternativemeasureshowsa30-districtdecrease.Similarly,whereasthestatisticsreportedinFigure3showthenarrowerrangehasatwelve-districtdecreaseincompetitivedistricts,thealternativemeasurehasathirteen-districtdecrease.

Page 17: Redistricting and Polarization

17

Second Order Connection: Geographic Sorting → District

Partisanship

Redistrictingcanmanipulatedirectlyonlydistrictpartisanship,buttowhatextentis

manipulationpossible?Instateswithasmallnumberofdistricts–orjustone–

manipulationofdistrictpartisanshipisneighimpossible.Evenwithinlargerstates,

scholarscontestwhetherredistrictingcanaffectdistrictpartisanship.Theargument

forminimalredistrictingeffectsarisesfromresearchthatarguesAmericanshave

residentiallysegregatedthemselves(BishopandCushing2009).Suchgeographical

sortingproduces,sotheargumentgoes,anincreasingpoliticalhomogenizationof

congressionaldistricts(Abramowitzetal.2006)thatvirtuallylocksinRepublican

majoritiesabsentDemocraticgerrymandering(ChenandRodden2014).The

correlationbetweendistricts’partisanshipandmembers’ideologyevidentinFigure

2thusemergesorganicallyfromstates’geographies,notthroughconscious

redistrictingmanipulation.

Evidenceforthis“bigsort”thesisisderivedfromcounty-levelpresidential

electionresultsfrom1976to2004,whichshowthatthenumberoflandslide

countieswithavictorymarginof20percentagepointsormoreincreasedfrom27

percentto48percentduringthisperiod(BishopandCushing2009).Scholarshave

challengedthesefindings.KlinknerandHapanowicz(2005),takingalongerview,

showsubstantialup-and-downvariationinthenumberoflandslidecounties

between1976and2004,withtherisebetweenthesetwoelectionsfallingwithinthe

historicalrangefrom1840to2004.AbramsandFiorina(2012)objecttothe

Page 18: Redistricting and Polarization

18

relianceonpresidentialelectionresults,notingthatpresidentialcandidatesarenot

allthesameand,thus,polarizationmeasuredbypresidentialelectionresultsmaybe

anartifactofthechoicesvotersarepresentedwith;4evenifvotersremainedthe

same,twomoderatecandidatesmayproducedifferentgeographicaldistributionof

votesthantwoextremecandidates.Utilizinganalternatemeasurefreerfrom

transitorycampaigneffects–partyregistration–therewasadecreaseinlandslide

countiesfrom1976and2004(McGeeandKrimm2009).

ThestatisticspresentedinFigure3alsodonotconformtothebigsortthesis,

althoughwedonotwishtotaketoomuchfromstatisticsdrawnfromdifferent

presidentialelections.Therewasaslightincreaseinthenumberofcompetitive

districtsbetween2002and2010,inboththewiderandnarrowerranges.Ifongoing

geographicalsortingofpartisansisprimarilyresponsibleforadeclinein

competitivedistricts(Abramowitzetal.2006),thereshouldhavebeenadecreasein

thenumberofcompetitivedistrictsoverthelastdecade.

Second Order Connection: Regional Realignment →

District Partisanship

InFigure1wedrawanarrowfromregionalrealignmenttopartydivergence,but

thepartydivergencetheoryimplicitlyinvolvesdistrictpartisanship.AsJacobson

notes(2013:548)“overthelastseveraldecades,changesinthepreferences,

behavior,anddistributionofcongressionalvotershavegiventhecongressional

4Furthermore,AbramsandPope(2012)arguethatAmericansarenotpolarizedgeographically;however,thiscritiqueisoutsidethescopeofouranalysis.

Page 19: Redistricting and Polarization

19

partiesmoreinternallyhomogenous,divergentandpolarizedelectoralbases…The

mainsourceofthiselectoraltransformationwasthepartisanrealignmentofthe

South.”TherepresentationalanchorforSouthernDemocratsweretheir

conservativeconstituents.Ifmembershadrepresentedliberalconstituencies,as

Democraticmembersdidelsewhere,theyprobablywouldnothavebeenso

conservative,likewisesowiththeirRepublicanreplacements(ifregional

realignmentwouldhaveevenoccurred).Yet,partydivergenceisnottypically

presentedlayingbaretheseassumptionsasJacobsondoes,sowedrawonlydotted

arrowsfromregionalrealignmenttodistrictpartisanshiptopartydivergence.

Theregionalrealignmentstory,whenframedasaffectingdistrict

partisanship,hasapotentialinteractionwithredistricting.SouthernDemocrats

attemptingtostaveoffextinctionwerethwartedbytheGeorgeBush,Sr.

DepartmentofJusticefollowingthe1990redistricting.TheDepartmentofJustice

orderedSouthernDemocratic–controlledstategovernmentstomaximizethe

numberofAfrican-Americanmajoritydistricts,atacticthattheU.S.SupremeCourt

wouldlaterfindunconstitutionalintheShawv.Renocases.Butthedamagewas

done.GrofmanandHandley(1998)findthatwithoutenoughAfrican-American

constituentstobuoytheirelectoralfortunes,someDemocraticincumbents

succumbedtomountingpressuresofregionalrealignment.5

5Today,withDemocratic-controlledSouthernstategovernmentsreplaced

withRepublican-controlledstategovernments,theVotingRightsActinsomeinstancesforcedRepublicanstodrawDemocraticdistrictswheretheymayotherwisenotwishtodoso(Schotts2002).

Page 20: Redistricting and Polarization

20

Full Causal Path: Redistricting → District Partisanship →

Elite Polarization

Astheprecedingdiscussionshouldmakeplain,redistrictingcannotberesponsible

forallelitepolarization.Scholarsfindredistrictingmakesacontribution.Stonecash,

Brewer,andMariani(2003)examinechangingdistrictcompositionandconclude

thesechangesproducemorehomogeneousdistrictsthatcontributetopolarization.

However,thesescholarsdonotdecomposethefactorsdrivingchangingdistrict

demographics.Strongerdirectevidencethatredistrictingaffectspolarizationcomes

fromstudiesthatexaminedistrictschangedbyredistricting.Carsonetal.(2007:

878)findthat“redistrictingisoneamongmanyfactorsthatproduceparty

polarization.”Theirmethodexaminestheideologyofmemberselectedfrom

congressionaldistrictsthatchangedsubstantiallyfollowingaredistricting.They

findthatthesedistrictstendtobemoreextremethanothersintheirpartisanship,

andthatmemberselectedfromthesedistrictsalsotendtobemoreideologically

extreme.Usingasimilarapproachtoexaminecongressionaldistrictsthatwere

changedbyredistricting,Hayes,Hibbing,andSulkin(2010)alsofindmembers’

behaviorisresponsivetodistricts’changeintermsofhowtheyco-sponsorbills

whentheirdistrictbecomesmorecompetitiveandhowtheyvoteinspecificissue

domainsthatmaybeofimportancetocertainconstituents,suchasrural

constituentsandfarmpolicies.Membersmaymodifytheirbehaviorinanticipation

ofconstituencychangewroughtbyredistricting(Boatright2004),althoughitmay

Page 21: Redistricting and Polarization

21

takeseveralelectioncyclesformemberstofullyadjusttotheirnewconstituencies

(Stratmann2000).

Yet,therearelimitstoredistrictingeffects.Carsonetal.(2007:889)findthat

only60percent(or261of435)ofdistrictboundariessubstantiallychangedfrom

thebeginningtotheendoftheentiretimespanfrom1962to2002.Othersaremore

sanguineonthecontributionofredistrictingtopolarization.McCarty,Poole,and

Rosenthal(2009)arguethatCarsonetal.(2007)overstateredistrictingeffects

sincedistrictsundergoingchangearethosefoundinthemostpolarizedstates.6

McCartyetal.(2009:667)examinetwocausesofpolarization:theaveragedistance

betweentheideologiesofthetwopoliticalparties(i.e.,partydivergence)andthe

congruenceofideologyanddistrictpartisanship.Theycallthelatter“sorting”

(2009:669)inhowmembersmaybearrayedalongthecontinuumofdistrict

partisanship;inthiscontext,sortinginvokestheconceptthatmembersofthetwo

politicalpartiesmaybeincreasinglyelectedfromdistrictsmorecongruentwith

theirpartyaffiliation.Theauthorsconcludefromexaminationofdetrendedsorting

effectsacrossaredistrictingthatredistricting“canaccountfor,atmost,10–15%of

6Itisalsotruethatlargerstatesarewheremoredistrictschangevis-à-vis

apportionment,andwherevoterscanbemostsegregatedinwaysthatwouldcontributetopolarization.

Page 22: Redistricting and Polarization

22

polarizationsincethe1970s.”7Theseauthorsassertmoreforciblyinthepopular

pressthat“gerrymanderinghasnothingtodowithpoliticalpolarization.”8

McCarty,Poole,andRosenthal(2009:672)setahighbarforredistrictingto

affectmembers’ideologies,inthatmembers’behaviormusthavea“sharpincrease”

immediatelyfollowingaredistricting,eventhoughmembersmayadjusttheir

behaviorpriortoaredistricting(Boatright2004)orgraduallyadjusttheirbehavior

overseveralelectioncyclesafteraredistricting(Stratmann2000).Furthermore,

McCartyetal.’smodelisstatic.Redistrictingmaybeusedtoshoreupanincumbent

mismatchedwithhisorherdistrict(Jacobson2003),suchthatdistrict

constituenciesaremadetobettermatchmembers’ideologies,particularlymembers

ofapartywhorepresentdistrictsthatleantowardtheotherpartyastheseare

generallythemostmoderateofallincumbents(seeFigure1).Carsonetal.(2007)

directlyexplorethiseffect,whileMcCartyetal.(2009)inferitbetweenchangesin

theirestimatedeffectsfromoneCongresstothenext.Astaticmodelalsocannot

7Elsewhere,McCarty,Poole,andRosenthal(2009:673)saythatredistricting

accountsfor“lessthan25%oftheincreaseinpolarizationsince1973.”UsingtheirfigurespresentedinTable2(p.673),wecalculatetheeffectfromtheirOLSmodeltobe25.1percentand19.4percentfromtheirmatchingmethod.The10–15percentfigureappearstobeadetrendedestimatethatassumesauniformincreaseinpolarizationinCongressesseatedbetween1973and2003.Theirunspecifieddetrendingmethodmaywashoutacuriousfeatureoftheirestimates,thatpolarizationduetosortingdeclinespriortoaredistricting(consistentwithBoatright2004)andthenincreasessharplyinallbutoneofsixoftheirestimates.

8McCarty,Nolan,KeithT.Poole,andHowardRosenthal.2013.“GerrymanderingDidn’tCausetheShutdown,”BloombergBusiness.Retrievedfromhttp://www.bloombergview.com/articles/2013–10-09/gerrymandering-didn-t-cause-the-shutdown.

Page 23: Redistricting and Polarization

23

measureotherpotentialchanges.Iftheconditionalpartygovernmentmodel(Rohde

1991)holdsforhowrealignmentcontributedtoelitepolarization,vis-à-visan

increasinglyhomogeneouspartycaucuswillingtogivemorepowerfultoolsfor

leaderstoenforcepartydiscipline,thenitstandstoreasonthereverseistrue,too:

makingdistrictsmorecompetitivewouldleadtoamoreheterogeneousparty

caucusandaweakeningofleaders’powers.

Whilethesestudiesfindredistrictingaffectselitepolarizationtovarying

degrees,someauthorssuggestthatcreatingmorecompetitivedistrictscould,tothe

contrary,increasepolarization.Inthepresenceofcompetition,candidatesmustrely

moreheavilyonpartysupport,andthusmaybemoreresponsivetopartyleaders

whodemandpartydivergence.CoxandKatz(2002)notethatincumbency

advantageincreasedfollowingthereapportionmentrevolutioninthe1960s,when

districtsinitiallybecamelesscompetitive.Engstrom(2013)notesthatelite

polarizationhashistoricallybeenhigherwhendistrictsweremorecompetitive.

However,McDonald(1999)notesthatthecausalarrowmaybereversed,thatwhen

voterspolarize,anoptimalgerrymanderingstrategytotakeadvantageofaless

volatileelectorateistodrawdistrictsthatonlyappearmorecompetitive.

Can Redistricting Be Reformed to Reduce Polarization?

Theprecedingdiscussionprimarilyinvolvesobservationalstudiesofthefactorsthat

driveelitepolarization.Ifredistrictingcanaffectdistrictpartisanship,thentowhat

Page 24: Redistricting and Polarization

24

degreecanredistrictingreformleadtoagreaternumberofcompetitivedistricts

andareductioninideologicalpolarization?Thisquestioncanbeansweredthrough

twomethods.First,observationalstudiesofstatesthatadoptalternative

redistrictinginstitutions–particularlycommissions,whicharethepreferredreform

model(McDonald2007)–revealredistrictinginterventioneffects.Second,

alternativemappingbycomputersimulationorbyhumansrevealshowalternative

mapsrangeacrossquantifiablemeasuressuchascompactnessanddistrict

competitiveness.

Observationalstudiestendtofindnomeasurableeffectofredistricting

institutionsonpolarization.Abramowitzetal.(2006)findredistrictingcommission

statesdonothaveagreaternumberofcompetitivedistricts.RyanandLyons(2014)

similarlyfindcommissionstatesdonothaveagreaternumberofcompetitive

districts–whattheycall“bipartisandistricts”–orreducedelitepolarization.

ExaminingCalifornia,Kousser,Phillips,andShor(201425)bluntlyfind,“Neither

theCitizensRedistrictingCommissionnorthetop-twoprimaryhashaltedthe

continuingpartisanpolarizationofCalifornia’selectedlawmakersortheirdrift

awayfromtheaveragevoterineachdistrict.Ifanything,polarizationhasincreased

andthequalityofrepresentationhasdeclined.”

Ageneralproblemwiththeseanalysesisthattheydonotdeeplyconsiderthe

causalmechanismslinkingredistrictingcommissions’rulesandmembershipto

morecompetitivedistrictsormoderationofelitepolarization.Onlytwostates,

ArizonaandWashington,explicitlyhavearedistrictingcriteriontodraw

competitivedistricts(McDonald2007).California’scommissionisnotrequiredto

Page 25: Redistricting and Polarization

25

drawcompetitivedistricts,sothetheoryisunderdevelopedastowhythestate’s

newlyenactedcitizencommissionwouldreducepolarization.Likewise,Abramowitz

etal.(2006)andRyanandLyons(2014)paintallcommissionswithabroadbrush.

Wearenotsurprisedthatcommissionsareindistinguishablefromlegislaturesif

classificationofcommissionsincludespredominantlypartisanorbipartisan

commissionsthatmaybepoliticallymotivatedtodrawpartisanorbipartisan

gerrymanders,boththattheoreticallyshouldresultinnouncompetitivedistricts

(OwenandGrofman1988).

Simulationshelpdevelopingcounterfactualsastowhatoutcomesare

feasiblewithinastate.Therangeofoutcomesisprofoundlyaffectedbystates’

geographies.Idaho’sbipartisancommissionsimplycannotdrawoneofthestate’s

twocongressionaldistrictstobeDemocraticleaningorcompetitive,whereas

Arizona’scommissionhasmoreopportunitiestodrawboth,byvirtueofitslarger

sizeandmorebalancedpartisanship,butisuniquelyconstrainedbyLatinovoting

rightsconcernsandotherstateconstitutionalredistrictingcriteria(McDonald

2006b).Observationalstudiesbydesigntreatstatesasequalcases,andscholarscan

offeronlycrudecovariatesatbesttocontrolforstates’differingcharacteristics.

Analternativeassessmentmethodologyistosimulatewhatmayhappen

withinastatethroughalternativemapping.Thisapproachhasbeenusedwith

humanmapperstoexaminemotivesofthe1990sNorthCarolinaredistricting

(GronkeandWilson1999),toforecasttheeffectsofCalifornia’s2000sredistricting

reform(Johnsonetal.2005),toexaminechangingtherank-orderingofArizona’s

criteria(McDonald2006b),andtoexploretheuseofalternativecriteriainfive

Page 26: Redistricting and Polarization

26

Midwesternstates(McDonald2009),Virginia(AltmanandMcDonald2013),Florida

(AltmanandMcDonald2014a),andOhio(AltmanandMcDonald2014c).Scholars

havealsousedautomatedmethodstoexploretherangeofpotentialredistricting

plans.AutomationwasfirstproposedbyVickery(1961)andhasbeenusedto

explorealternativesinfourstatesinthe1960s(Nagel1965),toproberacial

gerrymanderingintentinSouthCarolina(Cirincione,Darling,andO’Rourke2000),

toprobenationallythelimitsofcreatingcompetitivedistrictsinthe2000s

(McCarty,Poole,andRosenthal2009),andtoprobethroughacasestudyof

Florida’sgeographicalconstraintsonpartisangerrymandering(ChenandRodden

2014).

Wecreatedthefirstopen-sourceautomatedredistrictingalgorithm,called

BARD(AltmanandMcDonald2011).Throughourexperiencewithautomation,we

cametoasimilarconclusionasNagel(1965:899):thatautomatedalgorithmsare

“usefulfortestingsomepolicyproposals”byrevealingalternativesfor

consideration.However,automatedalgorithmsaredeeplychallengedbythewell-

establishedtheoreticallimitsofintegeroptimization–whatredistrictingisinthe

abstract–andaresusceptibletosubtleimplementationbiases(Altmanand

McDonald2010).Inanutshell,redistrictingoptimizationalgorithmsarenot

guaranteedinamodest-sizedstatetofindtheglobaloptimumofascoringfunction

(i.e.,criteria)inafinitetime.Becausetherearepracticallyaninfinitenumberof

localoptima,thesealgorithmstendtobecometrappedinlocaloptimaandarethus

Page 27: Redistricting and Polarization

27

biasedinunknownways.9Alternativeredistrictingplansdrawnbycomputersor

humanscanilluminatetrade-offsamongredistrictingcriteria,buttheabsenceofa

redistrictingplanwithagivenscoringprofiledoesnotmeanthataplan,orahostof

similarplans,doesnotexist.

Someimplementationissuesariseoutofsimplificationsthatresearchers

maketotheirdata.Forexample,McCarty,Poole,andRosenthal(2009)employa

simulationapproachtoexplorehowgeographyconstrainsthenumberof

competitivedistricts.Thesescholarsdrawdistrictsoutofcounties.However,the

largestcountiesintheUnitedStatescanentirelycontainseveralcongressional

districts.Toaddressthisissue,thesescholarssubdividelargecountiesinto1,000

personblocks,and“assumethateachofthesecountyblocksisidentical”intermsof

theirpoliticalcompositiontothecounty(McCartyetal.2009:674).Thus,9,546

identicalblocksarecreatedforLosAngelesCounty,whereObamawon70.6percent

ofthetwo-partyvotein2008.McCartyetal.(2009:674)assertwithoutproofthat

this“homogeneityassumptionbiasestowardsfindingagerrymanderingeffect”

becausethesimulationalgorithmisunlikelyto“produceeitherveryconservativeor

veryliberaldistricts”(2009:675).Thisseemsoddtous:thealgorithmwillproduce

roughly13LosAngelesCountycongressionaldistricts;allwillhaveexactlythesame

70.6percentObamavoteshare.Similarpatternswillbeobservedinotherlarge

countiesthathavetwoormorecongressionaldistricts.McCartyetal.’s(2009)data

9ThisistruewhenimplementingalgorithmstomeetallU.S.legalredistrictingcriteria.Anautomatedalgorithmdoesexisttocreateequalpopulation,contiguous,andcompact(withaspecificcompactnessscoringmechanism)(Olson2010).

Page 28: Redistricting and Polarization

28

assumptionisfurtheratoddswiththebigsortthesisthatDemocratsgerrymander

themselvesbypredominantlylivinginlargecounties(BishopandCushing2009;

ChenandRodden2014).Tothecontrary,McCartyetal.’s(2009)county

homogeneityassumptionbiasesagainstfindingagerrymanderingeffect.

Sinceredistrictingalgorithmsmayhavesubtlebiasesthataffectconclusions,

wewishtoexplorewhetherhumanscandrawplansthatexploreabroaderrangeof

outcomesacrosscriteriaofsubstantiveinterest,suchasthenumberofcompetitive

districtsandcompactness.Whereashumansmaydrawmapswithbiases,our

intuitionisthathumanswillbegintheirmappingfromvariousstartingpointsand

employcomplexheuristicsthathelpthemavoidtrappingthemselvesinlocal

optima,whereasthesimpleheuristicsemployedthusfarresearchcannot.An

observationofMexico’sexperiencewithautomatedredistrictingdemonstrates

humansarecapableofbeatingasimulatedannealingoptimizationalgorithm

(Altmanetal.2014).WorkingwithoursoftwaredevelopmentpartnerAzavea,we

developedopen-sourceweb-accessibleredistrictingsoftwarecalledDistrictBuilder,

whichwedeployedinseveralstatestosupportadvocates’efforts(Altmanand

McDonald2014b).Florida’slegislaturedevelopedasimilaronlinetooltoenable

greaterpublicparticipation.Wepresenttheresultsfromthreestudiestodateon

Florida(AltmanandMcDonald2014a),Ohio(AltmanandMcDonald2014c),and

Virginia(AltmanandMcDonald2013).

InFigure4weplotalllegalredistrictingplansthatwerepubliclyavailablein

Florida,Ohio,andVirginia.Wedefinelegalplansasthosewhereallgeographyis

assignedtothecorrectnumberofcongressionaldistricts,thatthesedistrictshavea

Page 29: Redistricting and Polarization

29

populationdeviationofplusorminus1.0percentfromtheidealequi-populous

district(manyplanshavedistrictsthatdeviatebyoneperson),andthereisa

minimumnumberofminority-majoritydistricts.10Publiclyavailableredistricting

plansarethoseeithermadeavailablebythestatelegislatureordevelopedthrough

advocates’redistrictingcompetitionsthatwesupported.Weplottheaverage

compactnessofdistrictingplansalongthehorizontalaxis,suchthatmorecompact

districtshavehighervalues,11andthenumberofcompetitivedistrictsinthetighter

plusorminusfourpercentagepointrange,usingthe2008two-partypresidential

vote,normalizedtothecandidates’nationalvoteshares.Wedisaggregateprecinct

levelpresidentialvotetothecensusblockleveltoconstructourstatistics.This

approachreplacesMcCarty,Poole,andRosenthal’s(2009)county-level

homogeneityassumptionwithaprecinctlevelhomogeneityassumption(amuch

smallerunitofaggregation),asisconsultants’typicalpracticewhenconstructing

redistrictingdatabases(McDonald2014).Weidentifytheadoptedplanwithared

triangle;allotherplansareidentifiedbybluediamonds.

Figure4revealsthatitispossibletocreatemorecompetitivedistricts

withoutsacrificingcompactnessinFlorida,Ohio,andVirginia;indeed,theplanwith

themostcompetitivedistricts–andmanyotherplanswithagreaternumberof

10Weanalyzeplanswithatleastthreemajority-minoritydistrictsinFlorida

andoneinbothOhioandVirginia.11WeusetheSchwartzburgcompactnessmeasure.Wenormalizethevalues

sincethevaryinggeographyofthesestatesdoesnotmakecomparisonsofscoresacrossstatesmeaningful.

Page 30: Redistricting and Polarization

30

competitivedistricts–ismorecompactthantheadoptedplan.12InFlorida,the

adoptedplanhasonecompetitivecongressionaldistrictoutof27,whileaplan

existswitheightcompetitivedistricts.InOhio,theadoptedplanhaszero

competitivedistrictsoutof16,whiletheplanexistswitheightcompetitivedistricts.

InVirginia,theadoptedplanhastwocompetitivedistrictsoutofelevenwhileaplan

existswiththreecompetitivedistricts.Inall,theadoptedplanshaveatotalofthree

competitivedistrictsoutof55,whileplanswiththemaximumnumberof

competitivedistrictshaveacombined21competitivedistricts.Wedonotknowifit

ispossibletodrawmorecompetitivedistrictssincethosedrawingplansmaynot

havebeentryingtomaximizedistrictcompetitiveness;indeed,inOhioandVirginia,

manynon-legislativemapdrawersparticipatedincompetitionswherecompetitive

districtswereonecriterionamongothers.

Figure 4 Competitive Districts and Compactness; All Florida, Ohio, and Virginia Publicly Available Legal Redistricting Plans

12Thisistrueforothercriteriasuchasrespectforpoliticalboundariesand

partisanfairness;forspaceconsiderations,wedonotpresentscatterplotsoftheseothercriteria(AltmanandMcDonald2013;2014a;2014b).

Page 31: Redistricting and Polarization

31

0

1

2

3

4

5

6

7

8

9

0 0.2 0.4 0.6 0.8 1

0

1

2

3

4

5

6

7

8

9

0 0.2 0.4 0.6 0.8 1

0

1

2

3

4

0 0.2 0.4 0.6 0.8 1

Ohi

o V

irgi

nia

Num

ber

of C

ompe

titiv

e D

istr

icts

Compactness

Page 32: Redistricting and Polarization

32

Discussion: Can Redistricting Reform Solve Polarization?

Ouranalysisofalternativelegalredistrictingplansthatincludeminorityvoting

rightsdistrictsrevealsthat,comparedtotheadoptedplansthathaveone

competitivedistrictoutof55,atleast21competitivedistrictscanbedrawnin

Florida,Ohio,andVirginia.McCarty,Poole,andRosenthal(2009:678),employing

simulationswithcontiguityandcompactnessconstraints,findthat80competitive

districtsarepossiblenationwide(althoughtheauthorsdonotdefinetheir

measurementofacompetitivedistrict).Wecannotfullyreconcilethesetwo

analysessinceMcCarty,Poole,andRosenthalexaminetheentirecountryandour

scopecoversonlythreeofthemostcloselycontestedbattlegroundstates.

Furthermore,wedonotbelievethatoneshouldsimplyextrapolateourfindingsto

allstates;forexample,thereareclearlysmalluncompetitivestateswithoneortwo

districtswhereredistrictingcanhavelittleornoeffect.Still,webelievethatthe

opportunitiesforcompetitivedistricts,andareductioninelitepolarization,maybe

greaterthantheMcCartyetal.analysisindicates.

WhereweagreewithMcCarty,Poole,andRosenthal(2009)isthatnotall

congressionaldistricts,andlikelynotevenamajority,canbemadecompetitive,and

thatredistrictingisbutonepieceofagreaterpolarizationpuzzle.Further,even

wheredistrictscouldbemadecompetitive,whetherredistrictingreformefforts

Page 33: Redistricting and Polarization

33

leadstocompetitivedistrictsdependsonthemannerinwhichthereformis

implemented.ContrasttheimplementationofredistrictingreforminCaliforniawith

theimplementationofreforminArizona.

AsKousser,Phillips,andShor(2014)discuss,California’sreformeffortis

likelytocreatecompetitivedistrictsandtoaffectpolarizationonlybyhappenstance.

Californiahasnoexplicitrequirementforcompetitivedistricts,andthecommission

adoptedonlythreecongressionaldistrictswitha2008Obamatwo-partyvoteshare

between48percentand52percent.Incontrast,inArizona,thecommissionis

explicitlytaskedbythestateconstitutiontodrawcompetitivedistricts.Arizona’s

commissionproducedthreecongressionaldistrictsoutofnine,whichisaseemingly

difficulttaskwhenalsodrawingtwoheavilyDemocraticLatinodistrictsinthis

Republicanleaningstate(McDonald2006b).LumpingArizonaandCaliforniaunder

thesamerubricofredistrictingcommissionsis,perhaps,whyscholarsfindthese

differentinstitutionshavenoeffectondistrictcompetition(Abramowitz,Alexander,

andGunning2006;RyanandLyons2014).(Similarly,Ohioadvocates’redistricting

competitionbalanceddistrictcompetitionagainstothercriteriatoapositiveeffect

ondistrictcompetition.SeeFigure4andAltmanandMcDonald2014c).Ifreformers

wishtoaffectapoliticaloutcome,werecommendexplicitlyincorporatingthe

intendedoutcomeintotheredistrictingcriteria.

OuranalysisdiffersfromsomeotherssuchasMcCarty,Poole,andRosenthal

(2009)infindingthatredistrictingreformcanhaveasubstantiveeffecton

polarization.Redistrictingisamongthefewinputstoelitepolarizationthatareboth

measurableandreadilysubjecttopolicyintervention.Webelieveredistricting

Page 34: Redistricting and Polarization

34

institutionscanbedevelopedtoproactivelyincreasethenumberofcompetitive

districtsandthusreduceelitepolarization.

Itispossiblethatredistrictingreformtoincreasecompetitivedistrictsmay

havesecond-ordereffectsbeyondthedirecteffectsofelectingmoremoderate

members.Thesesecondaryeffectscouldactintheoppositedirection;wenotethat

increaseddistrictcompetitionishistoricallycorrelatedwithgreaterideological

polarization(Engstrom2013).However,wejudge(followingEngstrom)that

historicalcorrelationisaresultofstrongandpolarizedpartiesusingbiaseddistricts

withthinmarginsinattempttomaximizepartisanadvantage–andnotaresultof

competitivedistrictscausingpolarization(McDonald1999).Moreover,thereare

theoreticalreasonstobelievethatthesecondaryeffectswillfurtherdecrease

polarization.Forexample,iftheconditionalpartymodelofgovernmentiscorrect

(Rohde1991),injectingtheHouseofRepresentativeswithmoremoderateswill

empowerthemtoconstrainthepowerofpartyleaderstoenforcepartydiscipline,

whichisamongtheprimarydriversofelitepolarizationthroughpartydivergence.

Basedonthecurrentevidence,weconjecturethatthesecondaryeffectswillbe

neutralorpositive–butnotethattheseeffectshaveyettobesystematically

measuredintheliterature.

Werecognizethattherearemanyotherreasonstoembraceredistricting

reformandcompetitivedistricts.Withspecificregardstocompetitivedistricts,they

providegeneralelectionvoterswiththeopportunitytoholdmembersaccountable

astheirdistrictsarenotsoelectorallysafe,anditlikewiseallowsvoterstohold

partiesaccountableastheirmajoritiesmaybeatrisk.Memberswhorepresent

Page 35: Redistricting and Polarization

35

competitivedistrictsmaynotdrawstrongchallengersiftheyrepresenttheir

districtswell(bybeingideologicallymoderate);however,itispossiblethata

greaternumberofcompetitivedistrictscanleadtoagreaternumberofcompetitive

elections,andhigherassociatedcampaigncosts.Likewise,competitiveelections

mayresultinagreaternumberofvotersbeingunhappyaboutthewinner(Brunell

2008).Weareremindedthatanyelectoralreformisinherentlyabalanceof

competingvalues(Arrow1951).Properlydesignedredistrictinginstitutionsmay

ameliorateelitepolarization,althoughtheeffectmaynotbeaslargeasadvocates

hope.Aswithallreforms,theremaybeunanticipatedconsequences.Still,ifoneis

concernedaboutelitepolarization,redistrictingshouldbeareformoptiononthe

table.

References Abramowitz, Alan I., Brad Alexander, and Matthew Gunning. 2006. “Incumbency, Redistricting,

and the Decline of Competition in U.S. House Elections.” The Journal of Politics 68 (1): 75–88.

Abrams, Samuel J., and Morris P. Fiorina. 2012. “‘The Big Sort’ That Wasn’t: A Skeptical Reexamination.” PS: Political Science & Politics 45 (2): 203–210.

Abramson, Paul R., John H. Aldrich, and David W. Rohde. 2014. Change and Continuity in the 2012 Elections. Washington, DC: CQ Press.

Allen, Liz, Amy Brand, Jo Scott, Micah Altman, and Marjorie Hlava. 2014. “Credit Where Credit Is Due.” Nature 508(7496): 312–313

Altman, Micah, Eric Magar, Michael P. McDonald, Alejandro Trelles. 2014. “Mexico’s Experience with Automated Redistricting.” Presented at the 2014 American Political Science Association Conference, Washington, DC.

Altman, Micah, and Michael P. McDonald. 2014a. “Paradoxes of Political Reform: Congressional Redistricting in Florida.” In Seth McKee, ed., Redistricting in the Sunshine State, Gainesville: University of Florida Press, 000–000.

Altman, Micah, and Michael P. McDonald. 2014b. “Public Participation GIS: The Case of Redistricting.” Proceedings of the 47th Annual Hawaii International Conference on System Sciences, Computer Society Press.

Altman, Micah, and Michael P. McDonald. 2014c. “Redistricting by Formula: An Ohio Experiment.” Unpublished Manuscript.

Page 36: Redistricting and Polarization

36

Altman, Micah, and Michael P. McDonald. 2013. “A Half-Century of Virginia Redistricting Battles: Shifting from Rural Malapportionment to Voting Rights and Participation.” University of Richmond Law Review 47: 771–831.

Altman, Micah, and Michael P. McDonald. 2011. “BARD: Better Automated Redistricting.” Journal of Statistical Software 42 (5): 1–28.

Altman, Micah, and Michael P. McDonald. 2010. “The Promise and Perils of Computers in Redistricting.” Duke J. Constitutional Law and Public Policy 5: 69–112.

Ansolabehere, Stephen, James M. Snyder, Jr., Charles Stewart, III, 2001. “Candidate Positioning in U.S. House Elections.” American Journal of Political Science 45 (1): 136–159.

Arrow, Kenneth. 1951. Social Choice and Individual Values. New York: John Wiley. Bafumi, J., and M. C. Herron. 2010. “Leapfrog Representation: A Study of American Voters and

Their Members of Congress.” American Political Science Review 104 (3): 519–552. Baldassarri, Delia, and Andrew Gelman. 2008. “Partisans without Constraint: Political

Polarization and Trends in American Public Opinion1.” American Journal of Sociology 114 (2): 408–446.

Bishop, Bill, and Robert G. Cushing. 2009. The Big Sort: Why the Clustering of Like-Minded America is Tearing Us Apart. New York: Houghton Mifflin Harcourt.

Boatright, Robert G. 2004. “Static Ambition in a Changing World: Legislators’ Preparations for, and Responses to, Redistricting.” State Politics and Policy Quarterly 4 (4): 436–454.

Brady, David W., Hahrie Han, and Jeremy C. Pope. 2007. “Primary Elections and Candidate Ideology: Out of Step with the Primary Electorate?” Legislative Studies Quarterly 32 (1): 79–105.

Brunell, Thomas L. 2008. Redistricting and Representation: Why Competitive Elections Are Bad for America. New York: Routledge.

Cain, Bruce. 1985. “Assessing the Partisan Effects of Redistricting.” The American Political Science Review 79 (2): 320–333.

Carson, Jamie L., Michael H. Crespin, Charles J. Finocchiaro, and David W. Rohde. 2007. “Redistricting and Party Polarization in the US House of Representatives.” American Politics Research 35 (6): 878–904.

Chen, Jowei, and Jonathan Rodden. 2014. “Unintentional Gerrymandering: Political Geography and Electoral Bias in Legislatures.” Quarterly Journal of Political Science 8 (3): 239–269.

Cirincione, Carmen, Thomas A. Darling, and Timothy G. O’Rourke. 2000. “Assessing South Carolina’s 1990s Congressional Districting.” Political Geography 19 (2): 189–211.

Cox, Gary W., and Jonathan N. Katz. 2002. Elbridge Gerry’s Salamander: The Electoral Consequences of the Reapportionment Revolution. Cambridge: Cambridge University Press.

Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row. Erik J. Engstrom. 2013. Partisan Gerrymandering and the Construction of American

Democracy. Ann Arbor, MI: University of Michigan Press. Erikson, Robert S. and Gerald C. Wright. 1980. “Elections and Policy Representation of

Constituency Interests: The Case of the 1974 House Elections.” Political Behavior 2(1):91-106.

Fiorina, Morris, Samuel J. Abrams, and Jeremy C. Pope. 2010. Culture War? The Myth of a Polarized America, 3rd ed. New York: Pearson Longman.

Page 37: Redistricting and Polarization

37

Gelman, Andrew, and Gary King. 1994. “Enhancing Democracy through Legislative Redistricting.” American Political Science Review 88 (3): 541–559.

Grofman, Bernard, Andres Blais, and Shaun Bowler, eds. 2009. Duverger’s Law of Plurality Voting: The Logic of Party Competition in Canada, India, the United Kingdom and the United States. New York: Springer.

Grofman, Bernard, and Lisa Handley. 1998. “Estimating the Impact of Voting-Rights-Act-Related Districting on Democratic Strength in the U.S. House of Representatives.” In Bernard Grofman, ed., Race and Redistricting in the 1990s. New York: Agathon Press, 51–67.

Grofman, Bernard and Gary King. 2007. “The Future of Partisan Symmetry as a Judicial Test for Partisan Gerrymandering after LULAC v. Perry.” Election Law Journal 6 (1): 2–35.

Gronke, Paul, and J. Matthew Wilson. 1999. “Competing Redistricting Plans as Evidence of Political Motives: The North Carolina Case.” American Politics Quarterly 27 (2): 147–176.

Großer, Jens, and Thomas R. Palfrey. 2014. “Candidate Entry and Political Polarization: An Antimedian Voter Theorem.” American Journal of Political Science 58 (1): 127–158.

Hayes, Matthew, Matthew V. Hibbing and Tracy Sulkin. 2010. “Redistricting, Responsiveness, and Issue Attention.” Legislative Studies Quarterly 35 (1): 91–115.

Hirsch, Sam. 2003. “The United States House of Unrepresentatives: What Went Wrong in the Latest Round of Congressional Redistricting.” Election Law Journal 2 (2): 179–216.

Jacobson, Gary C. 2013. “Partisan Polarization in American Politics: A Background Paper.” Presidential Studies Quarterly 43 (4): 688–708.

Jacobson, Gary C. 2003. “Terror, Terrain, and Turnout: Explaining the 2002 Midterm Elections” Political Science Quarterly 118 (1): 1–22.

Johnson, Douglas, Elise Lampe, Justin Levitt, and Andrew Lee. 2005. “Restoring the Competitive Edge.” The Rose Institute of State and Local Government, Claremont McKenna College.

Klinkner, Philip A. and Ann Hapanowicz. 2005. “Red and Blue and Déjà Vu: Measuring Political Polarization in the 2004 Election.” The Forum 3(2).

Kousser, Thad, Justin Phillips, and Boris Shor. 2014. “Reform and Representation: A New Method Applied to Recent Electoral Changes,” Social Science Research Network, June 3. Retrieved from http://ssrn.com/abstract=2260083 or http://dx.doi.org/10.2139/ssrn.2260083.

Layman, Geoffrey C., Thomas M. Carsey, John C. Green, Richard Herrera, and Rosalyn Cooperman. 2010. “Activists and Conflict Extension in American Party Politics.” American Political Science Review 104 (2): 324–346.

Mansbridge, Jane, and Cathie Jo Martin, eds. 2013. Negotiating Agreement in Politics: Report of the Task Force on Negotiating Agreement in Politics. Washington, DC: American Political Science Association.

Mayhew, David R. 1974. “Congressional Elections: The Case of the Vanishing Marginals.” Polity 6(3): 295-317.

McCarty, Nolan, Keith T. Poole, and Howard Rosenthal. 2009. “Does Gerrymandering Cause Polarization?.” American Journal of Political Science 53 (3): 666–680.

McCarty, Nolan, Keith T. Poole, and Howard Rosenthal. 2006. Polarized America: The Dance of Ideology and Unequal Riches. vol. 5. Cambridge, MA: MIT Press.

Page 38: Redistricting and Polarization

38

McGhee, Eric and Daniel Krimm. 2009. “Party Registration and the Geography of Party Polarization.” Polity 41(3): 345-376.

McDonald, Michael P. 2014. "Calculating Presidential Vote in Legislative Districts." State Politics and Policy Quarterly 14(2): 196-204.

McDonald, Michael P. 2007. "Regulating Redistricting." PS: Political Science and Politics 40(4): 675-9.

McDonald, Michael P. 2009. Midwest Mapping Project. Fairfax, VA: George Mason University. McDonald, Michael P. 2006a. “Drawing the Line on District Competition.” PS: Political Science

and Politics 39(1): 91–94. McDonald, Michael P. 2006b. “Re-Drawing the Line on District Competition.” PS: Political

Science and Politics 39(1): 99–101. McDonald, Michael P. 1999. “Representational Theories of the Polarization of the House of

Representatives.” Legislative Studies Section Newsletter, Extension of Remarks 22 (2): 8–10.

Owen, Guillermo, and Bernard N. Grofman. 1988. “Optimal Partisan Gerrymandering.” Political Geography Quarterly 7 (1): 5–22.

Nagel, Stuart S. 1965. “Simplified Bipartisan Computer Redistricting.” Stanford Law Review 71: 863–869.

Olson, Brian. 2010. Redistricter: A Non-Gerrymandered Impartial Redistricting Program. Available: http:// code.google.com/p/redistricter/ (last visited Feb. 26, 2015)

Poole, Keith T. and Thomas Romer. 1993. “Ideology, ‘Shirking’, and Representation.” Public Choice 77(1): 185-196.

Poole, Keith T., and Howard Rosenthal. 2000. Congress: A Political-Economic History of Roll Call Voting. New York: Oxford University Press.

Rohde, David W. 1991. Parties and Leaders in the Postreform House. Chicago: University of Chicago Press.

Ryan, Josh M., and Jeffery Lyons. 2014. “The Effect of Redistricting Commissions on District Bipartisanship and Member Ideology.” Journal of Elections, Public Opinion and Parties 25 (2): 234–263.

Shotts, Kenneth. 2002. “Gerrymandering, Legislative Composition, and National Policy Outcomes.” American Journal of Political Science 45 (2): 398–414.

Stonecash, Jeffrey M., Mark D. Brewer, and Mark D. Mariani. 2003. Diverging Parties: Social Change, Realignment, and Party Polarization. Boulder, CO: Westview Press.

Swain, John W., Stephen A Borrelli, and Brian C Reed. “Partisan Consequences of the Post-1990 Redistricting for the U.S. House of Representatives.” Political Research Quarterly 51(4): 945-67.

Stratmann, Thomas. 2000. “Congressional Voting over Legislative Careers: Shifting Positions and Changing Constraints.” The American Political Science Review 94(3): 665-676.

Theriault, Sean M. 2006. “Party Polarization in the U.S. Congress: Member Replacement and Member Adaptation.” Party Politics 12(4): 483-503.

Tufte, Edward R. 1973. “The Relationship between Seats and Votes in Two-Party Systems.” The American Political Science Review 67 (2): 540–554.

Vickrey, William. 1961. “On the Prevention of Gerrymandering,” Political Science Quarterly 76(1): 105-10.