Reading Report 2

political science report and need the explanation and answer to help me learn.

The report on the Jake Grumbach article should be 2-3 pages long (single spaced) and should include the following:
What is the “puzzle” or question that the article is trying to explain/solve?
What is the author’s primary argument?
How does the author differentiate their argument from previous ones?
What evidence does they use? How does this evidence support the argument? Is it sufficient?
Do you find the findings and analysis convincing?
List and define (in your own words and in the context of the article) three key terms the author uses.
Describe two things you like about the article. Explain why.
Do you find the argument and evidence to be convincing? Explain
Discuss and explain at least one weakness that you identify in the article.
Note: each question should be in a paragraph format, not bullet points, you can number the questions.
Requirements: 2-3 pages long (single spaced)
AmericanPoliticalScienceReview(2023)117,3,967–984doi:10.1017/S0003055422000934©TheAuthor(s),2022.PublishedbyCambridgeUniversityPressonbehalfoftheAmericanPoliticalScienceAssociation.ThisisanOpenAccessarticle,distributedunderthetermsoftheCreativeCommonsAttributionlicence(http://creativecommons.org/licenses/by/4.0),whichpermitsunrestrictedre-use,distributionandreproduction,providedtheoriginalarticleisproperlycited.LaboratoriesofDemocraticBackslidingJACOBM.GRUMBACHUniversityofWashington,UnitedStatesTheTrumppresidencygeneratedconcernaboutdemocraticbackslidingandrenewedinterestinmeasuringthenationaldemocraticperformanceoftheUnitedStates.However,theUShasadecentralizedformoffederalismthatadministersdemocraticinstitutionsatthestatelevel.Using51indicatorsofelectoraldemocracyfrom2000to2018,Idevelopameasureofsubnationaldemocraticperformance,theStateDemocracyIndex.Ithentesttheoriesofdemocraticexpansionandbackslidingbasedinpartycompetition,polarization,demographicchange,andthegroupinterestsofnationalpartycoalitions.Difference-in-differencesresultssuggestaminimalroleforallfactorsexceptRepublicancontrolofstategovernment,whichdramaticallyreducesstates’democraticperformanceduringthisperiod.Thisresultcallsintoquestiontheoriesfocusedonchangeswithinstates.Theracial,geographic,andeconomicincentivesofgroupsinnationalpartycoalitionsmayinsteaddeterminethehealthofdemocracyinthestates.Americanstates,whichwereoncepraisedbythegreatjuristLouisBrandeisas“laboratoriesofdemocracy,”areindangerofbecominglaboratoriesofauthoritarianismasthoseinpowerrewriteelectoralrules,redrawconstituencies,andevenrescindvotingrightstoensurethattheydonotlose.—LevitskyandZiblatt,HowDemocraciesDie(2018,2)INTRODUCTIONTheTrumppresidencyhasgeneratednewconcernsaboutauthoritarianismanddemocraticbackslidingintheUnitedStates(Dionne,Ornstein,andMann2017;Gessen2016;Liebermanetal.2019).Centraltothiscontemporarydiscussionhasbeenthemeasurementofnationaldemocraticperformance.Prominentcross-nationalmeasuresofdemocracyfromtheVarietiesofDemocracyProject(V-Dem),BrightLineWatch,andFreedomHouse,whichhadoncerankedthecountryasagloballeader,showaU.S.democracyslippingtoward“mixedregime”or“illiberaldemocracy”status.Yettherehasbeenlesssystematicinquiryintosub-nationaldynamicsinAmericandemocracy.ThisiscuriousinlightofAmericanfederalism,acompara-tivelydecentralizedinstitutionalsystemthatgivesstategovernmentstheauthoritytoadministerelections,drawelectoraldistricts,andexertpolicepower.LouisBrandeiscalledthestates“laboratoriesofdemocracy.”ButstategovernmentshavealsobeenforcesagainstdemocracyintheUSor,inthewordsofLevitskyandZiblatt(2018,2),“laboratoriesofauthoritarianism.”StateandlocalgovernmentsdirectlyandindirectlyenforcedaracialhierarchyformostofU.S.history(DuBois1935;Foner1988).ManyscholarsdonotconsidertheUnitedStatesademocracypriortothenationalenforcementoftheVotingRightsActof1965(VRA)againststategovernments(King2017;Mickey2015)—enforcementmademoredifficultbytheU.S.SupremeCourt’sdecisionsinShelbyCountyv.Holder(2013)andBrnovichv.DemocraticNationalCommittee(2021).Troublingstoriesaboundinrecentyears,ofvotersuppression,ofgerrymandering,ofstatelegislaturestakingpowerfromincomingout-partygov-ernors,andoftheauthoritarianuseofpolicepowersagainstvulnerablecommunities.Buttherehasbeenlittleefforttosystematicallytracethedynamicsofdemocraticperformanceinthestatesduringthecon-temporaryperiod.Inthisarticle,IcreateanewcomprehensivemeasureofelectoraldemocracyintheU.S.statesfrom2000to2018,theStateDemocracyIndex.Using51indicatorsofelectoraldemocraticquality,suchasaveragepollingplacewaittimes,same-dayandautomaticvoterregis-trationpolicies,andfelondisenfranchisement,IuseBayesianmodelingtoestimatealatentmeasureofdemocraticperformance.Analysisofthemeasuresug-geststhatstategovernmentshavebeenleadersindemocraticbackslidingintheUSinrecentyears.1Ifindsimilartrendswhenusingbroadermeasuresthatcoveradditionalcomponentsofdemocracysuchasliberalismandegalitarianism.IthenusetheStateDemocracyIndextoinvestigatethecausesofdemocraticexpansionanddeclineinthestates.Prominenttheoriesinpoliticalsciencepointtopartisancompetition(Keyssar2000),ideologicalpolarization(Liebermanetal.2019),racialdemo-graphicchange,andthegroupinterestsofnationalpartycoalitions(HackerandPierson2020)asimpor-tantdriversofdemocraticchange.PartisanJacobM.Grumbach,AssociateProfessor,DepartmentofPoliticalScience,UniversityofWashington,UnitedStates,grumbach@uw.edu.Received:May25,2021;revised:November06,2021;accepted:August23,2022.Firstpublishedonline:December01,2022.1Ratherthanasharpbreakinregimetype,thisinvestigationasksaboutmoregranularchangestoAmericandemocracythatinsomewaysparallelscomparativeanalysisof“hybrid”regimesthatcombineelementsofdemocracywiththoseofauthoritarianismandoligarchy(e.g.,LevitskyandWay2010).967https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
competitioncanincentivizepartiestoincorporatenewvoters(HajnalandLee2011;Teele2018b)orgener-atebrinksmanshipandscorched-earthtactics(Lee2009).Polarizationerodesnorms(LevitskyandZiblatt2018)andincreasestheideologicalcostofone’spoliticalopponentstakingpower(McCarty,Poole,andRosenthal2006).Statusthreatstemmingfromlocalracialdemographicchangecanprovokeinstitutionalbacklashagainstthedemocraticinclu-sionofBlackandLatinoAmericansorofrecentimmigrants(e.g.,BoboandHutchings1996;Enos2017).Finally,nationalpartiesthatrepresentbusi-nesshaveeconomicincentivestoconstraindemoc-racy(Ziblatt2017).ThecontemporaryRepublicanPartyisacoalitionoftheverywealthy,somemajorindustries,andanelectoralbasemotivatedinnosmallpartbywhiteidentitypolitics(ParkerandBarreto2014).Thesegroupshaveincentivestolimittheexpansionoftheelectoratetonewvoterswithverydifferentracialattitudesandclassinterests,suggestingthatRepublicancontrolofstategovern-mentmightreducedemocraticperformance.Ishowtrendsinstatedemocraticperformanceandtestthepredictionsofthesetheorieswithadifference-in-differencesdesign.Acrossmeasuresandmodelspecifications,theresultsareremarkablyclear:Republicancontrolofstategovernmentreducesdem-ocraticperformance.ThemagnitudeofdemocraticcontractionfromRepublicancontrolissurprisinglylarge,aboutone-halfofastandarddeviation.MuchofthiseffectisdrivenbygerrymanderingandelectoralpolicychangesfollowingRepublicangainsinstatelegislaturesandgovernorshipsinthe2010election.CompetitivepartysystemsandpolarizedlegislaturesdomuchlesstoexplainthemajorchangesinAmeri-candemocracyinthecontemporaryperiod.More-over,althoughtheRepublicanPartyhascapitalizedonracialanimusinrecentelections(Sides,Tesler,andVavreck2018),racialdemographicchangewithinstates—whetheronitsownorinconjunctionwithRepublicancontrol—playslittleroleinstate-leveldemocracy.Theseresultspointtowardnationalparti-sandynamicsratherthanwithin-statefactorsasthedriverofdemocraticchange.AsRocco(2021,6)writes,“[w]hileunevensubna-tionaldemocracyispreferabletoasituationinwhichterritorialgovernmentsareevenlyundemocratic,theexistenceofundemocraticoutliersneverthelesshelpstounderminedemocracyasawhole.”JustasslaveryandJimCrowintheU.S.SouthaffectedthepoliticsandsocietyoftheNorth,democraticbackslidinginstateslikeNorthCarolinaandWisconsinaffectsotherstates,and,moreimportantly,democracyintheUnitedStatesasawhole.Stateauthoritiesadministerelections,theyaretheprimaryenforcersoflaws,andtheydetermineinlargepartwhocanparticipateinAmericanpoliticsandhow.Thepolicyandjudiciallandscapeshavegrownincreasinglyfavorableforpolicyvariationacrossstatesinrecentyears.Asaconsequence,statesmaybeincreasinglyimportanttotrendsindemocracyacrossallinstitutionswithinAmericanfederalism.Politicalscholars,observers,andparticipantsshouldpaycloseattentiontodynamicsinstatedemocracy.MEASURINGDEMOCRACYINTHEU.S.STATESArichliteraturehasinvestigatedthebehaviorofU.S.stategovernments.Oneimportantareaoffocushasbeentherelationshipbetweenpublicopinionontheonehandandstatelegislativevotesandpolicyoutcomesontheother(CaugheyandWarshaw2018;Erikson,Wright,andMcIver1993;FlavinandFranko2017;Gay2007;LaxandPhillips2009;2012;Pacheco2013;Rogers2017;Simonovits,Guess,andNagler2019),includingwhetherstategovernmentalrespon-sivenesstothemasspublicisaffectedbytheinfluenceofconcentratedinterestgroupsandwealthyindividuals(Anzia2011;Hertel-Fernandez2014;RigbyandWright2013).Anadditionallargebodyofresearchhasaskedhowstateelectoralpoliciesaffectparticipa-tion(e.g.,Burdenetal.2014;Gerber,Huber,andHill2013).Thesestudieshaveaddressedcriticalquestionsofdemocracyinthestates,especiallywhetherstatepolicyoutcomesareresponsivetoandcongruentwiththepolicyattitudesofcitizens.However,therehasbeenlessquantitativestudyintowhystategovernmentsexpandorrestrictdemocracy—whytheymaketheirelectionsmoreorlessfreeandfair,andwhytheyexertauthorityinmoreorlessrepressiveways.2Thereisalsoaliteratureontheexistenceof“authoritarianenclaves”withindemocraticcountries(e.g.,Benton2012;Gibson2013),whichare“charac-terizedbyanadherencetorecognizablyauthoritariannormsandproceduresincontrasttothoseofthe[national]democraticregime”(Gilley2010,389).TheconceptofauthoritarianorundemocraticenclaveswithinpartlyorfullydemocraticcountriesisalsoseeninhistoricalresearchontheroleoftheU.S.statesinraciallyauthoritarianandundemocraticgovernance(King2017;Kousser1974;Mickey2015).DespitesuchimportantadvancesinthecomparativeandAmericanpoliticaldevelopmentliteratures,thereislittleinthewayofsystematicquantitativemeasure-mentofsubnationaldemocraticperformance(butseeHill1994).ConceptualizingDemocracyComponentsThisstudyfollowstheconceptualandmeasurementstrategiesofcomparativecross-nationaldemocracyresearch(e.g.,GleditschandWard1997;Lindbergetal.2014).Conceptualizingdemocracytofacilitatedifferentiation,whileavoiding“conceptualstretching”(Sartori1970,1034),is,ofcourse,challenging.Inconceptualizingandoperationalizingdemocracy,Ifol-lowscholarsinseparatingtheconceptintosubcompo-nents.Thisarticlefocusesmainlyonthesubcomponentofelectoraldemocracy.3Electoraldemocracycaptures2Anexceptionisinthestudyofstategovernmentalactionwithrespecttothepoliticalinclusionofnewimmigrants(GulasekaramandRamakrishnan2015).3Conceptualizingelectoraldemocracyasasubcomponentofdemoc-racyisdistinctfromitsconceptualizationasa“diminishedsubtype”ofdemocracyinsomecomparativeresearch(CollierandLevitsky1997,439).ElectoraldemocracyasadiminishedsubtypeimpliesthatJacobM.Grumbach968https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
whetherapoliticalsystemhaselectionsthatarefree,fair,andlegitimate,anditiscentraltohistoricalcon-ceptualizationofdemocracy(Dahl2003;Schumpeter1942).Theantithesisofelectoraldemocracyisautoc-racy,butIconceptualizeelectoraldemocracyasacontinuousratherthanbinarydimension.AnimportantnormativeandconceptualbasisforelectoraldemocracycanbefoundinDahl’s(1989)discussionof“polyarchy.”Thenecessaryconditionsforpolyarchy,whichLindbergetal.(2014)usetodevelopandmeasuretheirowncross-nationalconceptofelectoraldemocracy,includeelectedofficials,freeandfairelections,inclusivesuffrage,therighttorunforoffice,andadditionalinstitutionalcharacteristicssuchasassociationalautonomyandfreedomofexpression.Althoughallofthesecharacteristicsareimportanttothisparticularstudy,themostimportantcharacteristicsthatvaryacrossstatesinthecontemporaryperiodarefreeandfairelections—whethermembersofthepolityhaveanequalabilitytoinfluenceelectoral(and,byextension,policy)outcomes—andinclusivesuffrage—whethermembersofthepolityhaveequaleligibilityandaccesstotheballot.MostliteratureinAmericanpolitics,includingonstatepolitics,arguesthatcorrespondencebetweenpub-licopinionandpolicyoutcomesisanimportantindica-torofelectoraldemocracy.Incongruentorunresponsivepolicyoutcomesaresignsofa“demo-craticdeficit”(CaugheyandWarshaw2018;LaxandPhillips2012).However,IwishtonotethetensionsofWollheim’sparadox(Wollheim2016),inwhichalegit-imatedemocraticmajoritysupportsanundemocraticpolicy.Insuchasituation,isit“democratic”toimple-mentanundemocraticpolicy,suchasthedisenfran-chisementofaminoritygroup,accordingtothemajoritywill?Thisparadoxisrelevanttocontemporarypolicydebates,assurveysfindthatcertainvotersup-pressionpoliciesreceivemajoritysupportfromtheAmericanpublic(e.g.,Stewart,Ansolabehere,andPersily2016).Furthermore,thetheoreticaltraditionofBurkeanrepublicanismproposesamodelofrepre-sentationinwhichpoliticiansare“trustees”ofthepublicinterestwhoshouldactontheirownbeliefs,incontrasttothe“delegate”modelinwhichrepresenta-tivesshouldberesponsivetoconstituentopinion(MillerandStokes1963).Inmymeasures,IattempttobalancebothsidesofWollheim’sparadox,consider-ingpolicyresponsivenesstopublicopinionaswellasthecostofvoting,partisanbiasindistricting,andothernon-opinion-baseddimensionstobeimportantfordemocraticperformance.AsIaddressinAppendixSectionA5,scholarsacrossdisciplines(includingtheV-Demteam)haveconcep-tualizedadditionalimportantsubcomponentsofdemocracysuchasliberalism,egalitarianism,delibera-tion,andinclusion(forexamples,seeMichener2018;Mills2017;Phillips1991).Ibelievethatabroaderdefinitionofdemocracywouldincludethesecompo-nents.IprovidetwocorrespondingmeasurementextensionsintheAppendix,whereIcreateandanalyzebroadermeasuresofdemocracyinthestates.Thefirstextensionincludestheadditionalcomponentofliberaldemocracy.Liberaldemocracycaptureswhetherasoci-etyprotectscivilrightsandliberties(Brettschneider2010;Estlund2009),especiallyforminoritypopula-tionswhohavebeenhistoricallysubjugated(Glaude2017;Shelby2005).Liberaldemocracycanbecon-trastedwithauthoritarianism.Animportantinsightofrecentresearchhasbeenthecentralroleofthecarceralstate,whetherthestaterepressesitscitizenrythroughauthoritarianpolicingandmassincarceration,inshap-ingdemocraticperformance(SossandWeaver2017).Coercivestateauthority,seeninextremeformsinauthoritarianpolicingandmassincarceration,arealsomostlyadministeredwithstate-levelauthority(Miller2008;SossandWeaver2017;WeaverandProwse2020).Liberaldemocracymayalsoincludeconsider-ationsoftransparencyofdecisionmakingandpolicyinformation(Shapiro2009),and,empirically,democ-raciesaremoretransparentthanarenondemocracies(Hollyer,Rosendorff,andVreeland2011).Importantly,liberaldemocracyisconceptuallydis-tinctfrom“policyliberalism”(CaugheyandWarshaw2016),“sizeofgovernment”(Garand1988),andotherconceptsthatcapturetheleft–rightorientationofpol-icyoutcomesacrosspoliticalsystems.Onemightworrythatideologicalandpartisanconsiderationsinfluencethedefinitionofdemocracy,whichwouldleadtoatautologicalstudyofthecausesofdemocraticchanges.However,themainmeasureinthisstudy,withafocusonelectoraldemocracy,isnarrowlydefined.InthebroaderdemocracymeasureusedintheAppendix,theindicatorsofliberaldemocracyarealsocircum-scribedmorenarrowlythanthoseoftenfoundincom-parativedemocracyresearch(e.g.,Lindbergetal.2014).Furthermore,definingdemocracysoastoensurethedefinitionisbipartisanputsdemocracyresearchatgreaterriskoftautologyandthe“argumentfrommiddleground”fallacy,or,incontemporarypar-lance,“bothsiderism.”InasecondextensionintheAppendix,Icreateameasurethatcombineselectoral,liberal,andathirdcomponent,egalitariandemocracy.Tovaryingdegrees,scholarshaveaddressedcritiquesoftheconceptsofelectoralandliberaldemocracybyemphasizingequal-ityofrightsunderlaw—andtherealizationofrightsinpractice.Thesedebateshelpedtoconceptualizeanegalitariancomponentofdemocracythatfocusesonmaterialandsocialequalitybetweenindividualsandrelevantsubgroupsinthepolity(e.g.,Brettschneider2010;Przeworski1986).ThemultitieredfederalinstitutionalstructureoftheUSpresentsanadditionalconceptualchallengetoapolityhasfree,fair,andlegitimateelectionsbutlacksothernecessarycomponentstomakeita“full”democracy,suchascivilliberties,muchlikethediminishedsubtypeconceptof“maledemocracy”contrastswithpolitiesthatextenddemocraticcitizenshiptobothmenandwomen.Thus,theconceptualizationofelectoraldemocracyasasubcomponentmeansittakesus“up”theladderofgenerality(Sartori1970),applyingtomorecases,whereasitscon-ceptualizationasadiminishedsubtypetakesus“down”theladderofgenerality.LaboratoriesofDemocraticBacksliding969https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
investigatingthedemocraticperformanceofstates.ThisideaisrelatedbutnotidenticaltowhatGibson(2005,103)hasdescribedasthepotentialfor“anauthoritarianprovinceinanationallydemocraticcountry”(seealsoGibson2013).Notonlyarestatesnotseparate,atomizedpolitiesfromeachotherhori-zontally;theyareembeddedincomplexrelationshipswiththefederalgovernmentverticallyinastructureresemblingmoreofa“marblecake”thanthe“layercake”ofclassicaldualfederalism(Weissert2011).Theparticularwaythecakeismarbledisalsoinflux,changingdynamicallybasedonthepreferencesofcoa-litions(Riker1964;1975).Morespecifictothisarticle’sinquiryintodemocracy,stategovernmentsmayactinwaysthatexpandorcontractdemocracy,butonlydependentonfederalactivity.Forexample,theSupremeCourtinShelbyCountyv.Holder(2013)struckdowncriticalprovisionsoftheVotingRightsAct,allowingstatestoimplementchangestoelectoralproceduresinwaysthatthreatenthefreenessandfairnessofelections.4DemocracyIndicatorsInowturntomydemocracyindicators,theindividualvariablesthatIaggregateintotheStateDemocracyIndexmeasure.ForthemainStateDemocracyIndex(i.e.,electoraldemocracy)measure,Iuse51indicators.Atalevelbetweentheindicatorsandtheelectoraldemocracycomponent,theindicatorsfitintofmymeso-levelcategories:gerrymandering(e.g.,theparti-sanefficiencygap),electoralpoliciesthatincreaseordecreasetheeligibilitytoorcostofvoting(e.g.,felondisenfranchisementlaws),electoralpoliciesthatincreasetheintegrityofelections(e.g.,requiringpost-electionaudits),andobserveddemocraticoutcomes(e.g.,policyresponsivenesstopublicopinionandwaittimesforin-personvoting).Importantly,theStateDemocracyIndexcombinesindicatorsthatcapturedejureelectoralpoliciesandprocedures,whereasothersmeasuredemocraticoutcomeslikepolicyresponsive-nesstopublicopinionandvotingwaittimes.Together,theseindicatorscapturealargeamountofinformationrelatedtothefreedom,fairness,andequalityofvoiceinU.S.elections.Dataonsame-dayvoterregistration,earlyvoting,voterIDlaws,youthpreregistration,andno-faultabsenteevotingarefromGrumbachandHill(2021),anddataonautomaticvoterregistrationisfromMcGhee,Hill,andRomero(2021).Felondisenfran-chisementandprisonervotingpolicieswerecollectedfromtheNationalConferenceofStateLegislatures.Additionalelectoralvariables,especiallyvotingwaittimesandotherindicatorsofstateadministrativeper-formanceinelections,arefromtheMITElectionLab.5Gerrymanderingdata,whichfeatureprominentlyinthedemocracyindices,areprovidedbyStephanopou-losandWarshaw(2020),withanadditionaldistrictcompactnessmeasurefromKaufman,King,andKomi-sarchik(2019).6Ialsouseindicatorsofpolicyrespon-sivenesstopublicopinion(separatedintosocialandeconomicpolicydomains)basedonthestatepolicyandmasspublicliberalismmeasuresfrom(CaugheyandWarshaw2018).7Ilistall51indicatorsandtheirsourcesinAppendixTableA1.ForthealternativemeasuresusedinanalysesintheAppendix,Iuseindicatorscoveringliberaldemocracyandegalitariandemocracy.Theliberaldemocracyindi-catorscanbeputintothreemeso-levelcategories,withafocusonvariationinauthoritarianismthroughthecarceralstate(seeSossandWeaver2017):criminaljusticepolicies(e.g.,ThreeStrikeslaws),carceralout-comes(e.g.,theincarcerationrate),andcivillibertiespolicies(e.g.,protectionsforjournalistswithanony-moussources).IndicatorsrelatedtocriminaljusticearefromtheCorrelatesofStatePolicyDatabase(JordanandGrossmann2016)aswellastheBureauofJusticeStatisticsandInstituteforJustice.IalsoincludestateassetforfeitureratingsbytheInstituteforJustice“PolicingforProfit”dataset.8Indicatorsofegalitariandemocracyareinfivemeso-levelcategories:reproduc-tiverights,rightsforracialminorities,rightsforsexualminorities,welfarestateprovisions,andobservedsocioeconomicequality.TheStateDemocracyIndexcoverstheyears2000through2018.Ontheonehand,theshortnessofthisperiodisalimitation.Variationinelectoraldemocracyacrossstatesinthecontemporaryperiod,whichisthefocusofthisarticle,ismuchsmallerthanvariationduringtheslaveryandJimCrowperiods.However,throughvoterregistrationrules,electionadministra-tionprocedures,andlawsthatunequallyincreasethecostofvoting,statesstillvaryconsiderablyinhowinclusivesuffrageis.States’gerrymanderingoflegisla-tivedistrictboundarieshasalsogeneratedvariationinhowfreeandfairelectionsare,expandinginequalityinhowmuchindividuals’votesinfluenceelectionout-comesandreducingthepotentialformajoritarianrule.Furthermore,thereareseriouschallengestocreatingameasurethatdirectlycomparesinterstatevariationindemocracyinthecontemporaryperiodwiththatofearliereras,suchastheJimCrowperiod.9BylimitingtheStateDemocracyIndextothepasttwodecades,4QuantitativestudiesbuttresshistoricalresearchshowingthattheVotingRightsActhadprofoundeffectsonlegislativeresponsivenesstoBlackvoters(SchuitandRogowski2017)andonracialinequalityinlabormarketoutcomes(AnejaandAvenancio-León2019).5Availableatelectionlab.mit.edu/data.Idonotincludeavoter-turnoutvariableinthemeasurebecauselowturnoutcouldbeasignofdemocraticproblems(e.g.,adeficitofpoliticalefficacyandinclusionamongcitizens)ordemocratichealth(e.g.,citizenswhoapproveofthestatusquo;Lipset1960,chap.7).6Indicatorsofgerrymanderingthatmeasureoneofthetwoparties’advantage(e.g.,efficiencygap)aretransformedintotheirabsolutevaluestomeasuretheextentofpartisanadvantageineitherdirection.7Specifically,Iusethesquaredresidualsfromabivariateregressionofstatepolicyliberalismonstateopinionliberalism,whichcapturehow“outofstep”astate’spolicyiswithitsresidents’policyattitudes.8Availableathttps://ij.org/report/policing-for-profit-3/policing-for-profit-data/.9Thischallengeissimilartoestimatingthemedianlegislators’idealpointcivilrightsinthepre-andpost-civilrightseras.Post-1960slegislativecontestationwasoveramuchsmallerrangeoftheJacobM.Grumbach970https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
IbothcaptureaneraofimportantcontestationoverAmericandemocracyandavoidbridgingbetweenperiodsforwhichthereisverydifferentdataavailabil-ity,and,moreimportantly,potentiallyincomparabletermsofcivilandhumanrights.MeasurementModelsForthemainStateDemocracyIndexmeasure,Imodeldemocracyasalatentvariable(TreierandJackman2008).Thislatentvariableanalysisletsobservedrela-tionshipsbetweenthedemocracyindicatorsdeterminehoweachindicatorshouldaffectstates’democracyscores.Thisstrategyestimatesan“idealpoint”onalatentdimensionforeachstate-yearthatbestpredictsthevaluesofdemocracyindicatorsintheobserveddata.Inparticular,IuseBayesianfactoranalysisformixeddata(Quinn2004)becausethedemocracyindi-catorsmaybebinary(e.g.,same-dayvoterregistra-tion),ordinal(e.g.,disenfranchisementofall,some,ornofelons),orcontinuous(e.g.,legislativedistrictefficiencygap).Themodelisbasedontheequationbelow.Thedistributionofdemocraticperformanceonindicatorsforstatesinyeart,y∗st,isafunctionofthestate’slatentdemocraticperformanceforthatyear,θst,aswellasthedemocracyindicator’sdiscriminationparameterβjanddifficultyparameterαj.10Subscriptjdenotesdifferentindicators,whichareanalogoustotestquestionsintheitemresponsetheoryframework.Inthisequation,Njisanormaldistributionwithjdimensions(astherearejindicators)andΨisaJJvariance-covariancematrix.y∗stNjβjθst−αj,Ψ:(1)Themainbenefitofthisfactoranalysisisthatthemeasurerequireslittleinthewayofassumptionsfromusabouthowanyparticularindicatorshouldaffectdemocracyscores.11However,thiscomesatthecostofsomelossofcontrol;insomecircumstances,theestimatedparametersfordemocracyindicatorscanbe“wrong”intheoreticalandsubstantiveterms.Whetherornotyouconsiderthisaseriousproblemisdependentonwhetheryouphilosophicallyinterpretthese“errors”asmeasurementerrororbias.12Inaddition,theBayes-ianfactoranalysismodelprovidesestimatesofuncer-taintyforparameters(bothstatedemocracyscoresanddemocracyindicatoritemparameters).Figure1showsthediscriminationparameteresti-mates,βjfordemocracyindicatorj:Inshort,thedis-criminationparametersrepresenttheslopeoftherelationshipbetweenanindicatorandastate’slatentdemocracyperformancescore.Indicatorswithpositivediscriminationparametersincreaseastate’sdemocracyscore,whereasitemswithnegativeparametersdecreasethem.13ThediscriminationparametersinFigure1sug-gestthatasmallnumberofindicatorsdonotloadwellontothelatentdemocracydimension(withdiscrimina-tionparametersclosetozero),suchasthenumberofmilitaryandoverseasballotsnotreturnedandrestric-tionsonvoterregistrationdrives.Overall,however,theitemdiscriminationparametersareconsistentwiththe-oreticalexpectationsandsuggestthatelectoraldemoc-racyisunidimensional.Whenitemparametersdonotconformtotheory,onesolutionistodirectlyimposeitemparametersontheindicatorsratherthanmodelthem.Todoso,inadditiontotheBayesianfactoranalysismeasure,Iusesimpleadditiveindexingtocreateanalternativedemocracymeasure.Intheadditiveindex,Iweighteachdemocracyindicatorequallybyrange,scalingeachtothe[0,1]interval,andthentakethestateaverageacrossalltheindicators.Policiesthataredemocracycontracting,suchasfelonydisenfranchise-ment,arereversecoded.Thisisequivalenttoaddingupallofastate’sdemocracy-expandingpoliciesandthensubtractingthesumofdemocracy-contractingpolicies(forapplicationsofthismethodtostatepolicyliberal-ism,seeErikson,Wright,andMcIver1993;Grumbach2018).IproviderobustnesscheckswiththisadditivemeasureintheAppendix,andtheresultsareverysimilartotheresultswiththedata-drivenBayesianmeasureusedinthemainanalyses.ItestthevalidityoftheStateDemocracyIndexindifferentways.Icheckconstructvalidationbycompar-ingmymeasuretomeasuresofrelatedconcepts.Tomyknowledge,theclosestanaloguetomymeasureistheCostofVotingIndex(COVI)fromLi,Pomante,andSchraufnagel(2018),whichisbasedonsevenstateelectoralpolicyvariablesinpresidentialelectionyears.Statedemocracy,asaconcept,isrelatedtothecostofvoting.Ithereforecheckmymeasure’sconvergentvaliditybyestimatingitscorrelationtothispreviousmeasureinFigureA1intheAppendix,findingamoderatelystrongcorrelationof-0.71(highervaluesofCOVIindicategreatercostofvoting).Ialsoshowthatmymeasureispositivelycorrelatedwithideologicalspacewhenitcomestocivilrights(CaugheyandSchickler2016).10Themodelrequireslimitingtheparameterspaceforasmallnumberofitems;Ifixfiveitemdiscriminationparameterstobepositiveornegativebasedontheoreticalinterpretation.Iranthemodelwith20,000Gibbsiterationsforthesampler,withaburn-inperiodof1,000iterations.Tomaintainaconstantsubstantiveinter-pretationofhow“democratic”agivenindicatorisacrosstime,Imodeltime-invariantdifficultyparametersincontrasttothepolicyliberalismmeasureofCaugheyandWarshaw(2016).11Bayesianlatentdimensionmodelslikethisonerequirethemod-elertoconstraintheparameterspace.Idothisbyassigningarandomsetoffiveindicatorsapositiveornegativedifficultyparameterbasedonwhetheritistheoreticallydemocracyexpandingorcontracting(forasimilarapplicationtostatepolicyliberalism,seeCaugheyandWarshaw2016).12Itisalsoworthnotingthaterrorinthesedemocracymeasureswillreducetheprecisionofhypothesistests,butbecauseIusethesedemocracymeasuresasdependentvariables,thiswillnotinducebiasorinconsistency(amongmanysources,seeAngristandPischke2008).13Notshownhere,difficultyparametersαjareinterceptsthatscaletherelationshipsbetweenindicatorsanddemocracyscores.LaboratoriesofDemocraticBacksliding971https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
state-levelturnoutofthevoting-eligiblepublicinFigureA2intheAppendix.Iunfortunatelyhavelittleopportunitytotestforconvergentvalidationbecauseofthelackofexistingmeasuresofoverallstate-leveldemocraticperformance.Thereisscholarlyinterestinmeasuringsubnationaldemocraticperformanceatthecountrylevel(seeGiraudy2015;McMann2018),andasmallnumberofquantitativemeasuresofdemocracywithinothercountries’politicalsubunits(Harbers,Bartman,andvanWingerden2019),butIhavenotfoundsuchameasureofdemocraticperformancefocusedontheU.S.states.Inthenextsections,Iinvestigatedescriptivetrendsinstatedemocraticperformanceandthenturntoexplainingthesetrendswiththeoriesbasedinpartycompetition,polarization,demographicchange,andthegroupinterestsofnationalpartycoalitions.TRENDSINSTATEDEMOCRACYWiththeStateDemocracyIndexinhand,Ifirstexplorevariationbetweenstates,andwithinstatesacrosstime,indemocraticperformance.Figure2showsamapofFIGURE1.FactorLoadingsofDemocracyIndicatorsGerrymandering: partisan symmetry (Cong.−Pres.)Gerrymandering: partisan symmetry (Cong.)Gerrymandering: mean−median difference (state leg.)Gerrymandering: declination (Cong.)Gerrymandering: mean−median difference (state leg.−pres.)Gerrymandering: partisan symmetry (state leg.−pres.)Gerrymandering: declination (Cong.−Pres.)Gerrymandering: partisan symmetry (state leg.)Gerrymandering: mean−median difference (Cong.−Pres.)Gerrymandering: Efficiency gap (Cong.−Pres.)Gerrymandering: Efficiency gap (Cong.)Opinion−policy difference (social)Gerrymandering: mean−median difference (Cong.)Gerrymandering: declination (state leg.)voter ID (strict)voter ID (any)voters deterred because of disability or illness (off−year)Gerrymandering: declination (state leg.−pres.)voting wait timesvoters deterred because of disability or illness (on−year)number of felons ineligible to vote as percent of state populationfelony disenfranchisementGerrymandering: Efficiency gap (state leg.)Opinion−policy difference (economic)website for precinct ballotDistrict compactnesswebsite for absentee statuswebsite for registration statusGerrymandering: Efficiency gap (state leg.−pres.)website with polling placedata completenessregistration or absentee ballot problems (on−year)registrations rejectedprovisional ballots rejectedRestrictions on voter reg. drivesunder− and over−votes cast in an electionmilitary and overseas ballots not returnedwebsite for provisional ballot checkmilitary and overseas ballots rejectedpercent of eligible voters who registerAutomatic Voter Registration (any)registration or absentee ballot problems (off−year)Automatic Voter Registration (back end)online registrationState allows currently incarcerated to voteearly votingYouth preregistrationpostelection audit requiredprovisional ballots castSame day registrationabsentee voting−0.8−0.40.00.4Discrimination ParameterVariableNote:ThefigurepresentsthediscriminationparameterestimatesandBayesiancredibleintervalsforindicatorsusedintheStateDemocracyIndex.JacobM.Grumbach972https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
statescoresintheyear2000(leftpanel)andintheyear2018(rightpanel).ThemapsinFigure2showsomeclearregionalvariation,especiallyin2018.StatesontheWestCoastandintheNortheastscorehigheronthedemocracymeasuresthandostatesintheSouth.NewMexico,Colorado,andsomeMidwesternstatesalsohavestrongdemocracyscores.Themapsalsoshowwithin-statechangeduringthisperiod.StateslikeNorthCarolinaandWisconsinareamongthemostdemocraticstatesintheyear2000,butby2018theyareclosetothebottom.IllinoisandVermontmovefromthemiddleofthepackin2000toamongthetopdemocraticperformersin2018.Figure3highlightsacaseofmajorchangeindemo-craticperformance,NorthCarolina.Althoughthestatewasnotoriouslydifficulttodemocratizeduringthecivilrightsperiod(Mickey2015)—itmaintaineditsJimCrowliteracytestsforvotinguntilthe1970s—NorthCarolinahadbecomealeaderinexpandingaccesstovotingduringthelate1990sandearly2000s.Thestatehadexpandedopportunitiesforearlyvotingaswellasimplementedpoliciestoexpandvoterregistration,suchassame-dayregistrationandpreregistrationforyouth.FIGURE2.DemocracyintheStates,2000and201820002018−3−2−101DemocraticPerformanceNote:LeftpanelshowsStateDemocracyIndexscoresfortheyear2000.RightpanelshowsStateDemocracyIndexscoresfortheyear2018.FIGURE3.TheWeakeningofDemocracyinNorthCarolina−2022000200520102015YearDemocracy ScoreStateNCTXWAOtherNote:LinesrepresenttheStateDemocracyIndexscoresforstates(2000–2018).ThesolidblacklinerepresentsNorthCarolina,thedashedlinerepresentsTexas,andthedottedlineWashington.ShadedribbonsareBayesiancredibleintervals.LaboratoriesofDemocraticBacksliding973https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
Voterturnouthadincreasedbyover10percentagepointsonaverageduringthistime.ButamajorshiftoccurredaftertheRepublicanPartywoncontrolofbothlegislativechambersin2010.Beginningin2011,NorthCarolinamadeaseriesofchangestoitselectionlawsandprocedures.Thestateredrewitslegislativedistrictboundaries.Thenewdis-tricts,whichreceivedrapidcondemnationfromDem-ocratsandcivilrightsgroups,clearlyadvantagedwhiteandRepublicanvoters.In2018,forexample,Republi-canswonabout50.3%ofthetwo-partyvoteinNorthCarolina—butthisbaremajorityofvotesfromtheelectoratetranslatedtofully77%(10of13)ofNorthCarolina’sseatsinCongress.Scholarsofgerrymander-ingsuchasChristopherWarshawhavecalledNorthCarolinadistricts“probablythemostgerrymanderedmapinmodernhistory.”14AfterelectingaRepublicangovernorin2012,theunifiedRepublicangovernmentthenimplementedastrictvoterIDlawandcurtailedearlyvotinglawsinareaswithheavierconcentrationsofBlackvoters.ThesechangesarereflectedinFigure3.Figure4showstrendsinstatedemocracybyparty,withthesolidlinerepresentingunifiedRepublicanstatesandthedashedanddottedlinesrepresentingDemocraticanddividedstates,respectively.Thestatespolarizebypartyoverthisperiod:theaveragedividedstateandDemocraticallycontrolledstatebecomemoredemocratic,whereastheaverageRepublican-controlledstatebecomeslessdemocratic.However,thegroupsofstatescontrolledbyeachpartychangeoverthisperiod;IdonotknowfromFigure4whetherRepublicanstatesarebecominglessdemocraticorwhetherlessdemocraticstatesarebecomingmoreRepublican.Thepartisanrelationshipscouldalsobeconfoundedbyotherpotentialcausesofdemocraticchanges:competitionandpolarization.EXPLAININGDYNAMICSINSTATEDEMOCRACYTheStateDemocracyIndexmeasuresdevelopedintheprevioussectionssuggestthattherehavebeenmajorshiftsindemocraticperformancewithinstatesinrecentyears.However,theimportantquestionisnotsimplyhowdemocracyhaschangedinthestatesbutwhy.Luckily,thenewdemocracymeasuresallowustotestthepredictionsofcompetingtheoriesofthecausesofdemocraticchanges.Whatdrivesdemocraticexpansionsandcontractionsinpoliticalsystems?Politicalscienceofferssomepoten-tialexplanations.Theexplanationsengagewithtrans-formativeprocessesinmodernAmericanpolitics:partisancompetition,ideologicalpolarization,andnationalpartygroupcoalitions.Scholarspointtotheconsolidationofacompetitivepartysystemtoexplainlarge-scaleexpansionsofdemocracyintheUS(Teele2018a),Africa(RaknerandVandeWalle2009),Europe(Mares2015),andaroundtheworld(Weiner1965).Partiesincompetitiveenvironmentsmighthaveincentivestoexpandtheelectorateinsearchofmorevotes,improvingdemocracyintheprocessby,forexample,expandingthefranchise(Keyssar2000;Teele2018b).Ontheotherhand,however,byincentivizingpartisanbrinksmanship(Lee2009),partisancompeti-tioncanleadapartywithaprecariousgriponpowertoFIGURE4.DemocracyintheStatesbyPartyControlofGovernment−3−2−10122000200520102015YearDemocracy ScoreRepublicanDividedDemocraticNote:PlotshowsaverageStateDemocracyIndexscoresforstatesunderunifiedDemocratic(dottedline),divided(dashedline),andunifiedRepublican(solidline)control.Shadedribbonsare95%confidenceintervals.14TweetonOctober30,2019.AccessedFebruary2021.https://twitter.com/cwarshaw/status/1189597322331734016?s=20.JacobM.Grumbach974https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
diminishdemocracybyexploitingcountermajoritarianinstitutionsandattemptingtopreventtheiropponents’electoralbasesfromvoting.Ifollowresearchthatusesmeasuresoflegislativeandelectoralcompetitionwithinstatesasprimaryexplanatoryvariables(O’Brian2019;Teele2018b).Asecondtheoryfocusesonpolarization—theideo-logicaldistancebetweentheparties’agendas.Polariza-tionincreasespoliticians’needtoensurethattheiropponentsdonotwinoffice.Apartyingovernmentinapolarizedstatewillthushavegreaterincentivetochangepoliciesthataffectdemocracy,suchaselectionlawsthatinfluencethecostofvotingfordifferentgroupsinthestate.AsLiebermanetal.(2019,471)argue,“hyperpolarizationmagnifiestendenciesforthepartisancaptureofinstitutionsthataresupposedtoexercisechecksandbalancesbutmayinsteadbeturnedintounaccountableinstrumentsofpartisanorincumbentadvantage.”It“erodesnorms”ofinstitutionalbehavior,suchasthejudicioususeofexecutivepowerandfairtreatmentonissuessuchasbureaucraticandjudicialappointments—andtheleversofdemocracy,itself(LevitskyandZiblatt2018).Polarizationmaybeasym-metricorsymmetric(HackerandPierson2005;McCarty,Poole,andRosenthal2006),butpolarizationisfundamentallyaboutthedistancebetweentheparties.Ifollowliteraturethatusesthedifferenceinpartymediansinstatelegislaturesasameasure(ShorandMcCarty2011).Athirdtheoreticaltraditionsuggeststhattheracialdemographicsofstatepopulationsshapespoliticsandpolicy(HeroandTolbert1996).Ofparticularimpor-tancetothisstudyisthepotentialforincreasingracialdiversitytogenerate“racialthreat”andbacklashamongconservativewhitevoters(BoboandHutchings1996).Asstatesgrowmoreraciallydiverseduetoimmigrationandinternalmigration,15somevotersmightdemandrestrictionsondemocracytoblockthepoliticalinclusionandempowermentofnewvotersofcolor(AbrajanoandHajnal2017;BiggersandHanmer2017;MyersandLevy2018).Importantly,racialbacklashwouldnotonlyleadtodemocraticbackslidingonitsown;ifdemographicchangeleadsvoterstoincreasinglyelectRepublicanstostategovernment,thistheorypredictsthattheinteractionofdemographicchangeandRepublicanPartycontrolshouldproducedemocraticbacksliding.Finally,asetoftheoriesfocusesnotoncompetition,polarization,ordemographicchangewithinstatesbutontheinterestsofgroupsinnationalpartycoalitions.Torinstance,Ziblatt(2017)pointstotheimportanceofconservativepartiesashistoricalcoalitionsofgroupswitheconomicincentivestoconstraindemoc-racy.ThemodernRepublicanParty,which,atitselitelevel,isacoalitionoftheverywealthy,hasincentivestolimittheexpansionoftheelectoratewithnewvoterswithverydifferentclassinterests(HackerandPierson2020).Inrecentyears,largefirmsandwealthyindividualshavemademajorpoliticalinvest-mentsatthestatelevel,providing“legislativesubsidies”intheformofmodelbills,lobbying,andorganization,asHertel-Fernandez(2019)showsinthecasesoftheAmericanLegislativeExchangeCouncil,AmericansforProsperity,andtheStatePolicyNetwork.Incontrast,theGOP’selectoralbaseisconsiderablylessinterestedintheRepublicaneconomicagendaoftop-heavytaxcutsandreductionsingovernmentspending.However,theirpreferenceswithrespecttoraceandpartisanidentityprovidetheRepublicanelec-toralbasewithreasontoopposedemocracyinadiver-sifyingcountry.(SurveyevidencefromGrahamandSvolik(2020)alsosuggeststhatAmericanvotershavelittleinterestinmaintainingdemocraticperformanceifitmeansconcedingtheirpartisanorpolicygoals.)Thepoliticsofracearethereforestillcentraltothistheoryofpartycoalitions.However,unlikethelocalizedracialandpoliticaleconomyconflictoftheJimCrowperiod,todayitisnationalratherthanstateorlocallevelracialconflictthatisthedriver.Furthermore,increasingeconomicinequalitysincethe1970shascausedtheeconomicinterestsofthoseatthetoptodivergefromthoseofthemedianvoter(MeltzerandRichard1981).Thisdivergenceincentiv-izeseconomicelitestoeithermoderatetheireconomicagenda—whichtheRepublicanPartyhasnotdone—ortoappealtoalternativedimensionsofpoliticalconflict(HackerandPierson2020),themostcontentiousofwhichintheUSisracebutcanalsoincludeconflictovergender,religion,sexuality,andculture.Overall,thistheorysuggeststhatthecurrentcoalitionalstructureofthenationalRepublicanParty,shapedinlargepartbytwentieth-centuryracialrealignment(Schickler2016)andlargepoliticalinvestmentsbywealthyindividualsandfirms(HackerandPierson2010;Hertel-Fernandez2019),makesthepartyingovernmentespeciallylikelytoreducestatedemocraticperformanceinanystateinwhichittakespower.Iamalsointerestedintheinteractionsofcompeti-tion,polarization,andRepublicancontrol.Polarizationmightonlymatterincompetitivecontexts,whentheideologicallydistantout-partyhasarealchanceoftakingpower.Similarly,Republicancontrolmightonlyleadtobackslidinginacompetitiveenvironmentwheretheyrisklosinglegislativemajoritiesandgovernor-ships.TheinteractionofpolarizationandRepublicancontrolmightproducebackslidingifbackslidingisbeingdrivenbythemostideologicallyextremeRepub-licanstatelegislatures.Furthermore,theinteractionofracialdemographicchangeandRepublicancontrolmightleadtobackslidingifgrowingminoritypopula-tionsprovokeracialthreatamongwhitevoters,leadingthemtoelectRepublicans,withagoalofstemmingtheexpandingelectoralpowerofminorityvoters.IcontinuethisdiscussionofthepotentialcausesofdemocraticexpansionandcontractioninAppendixSectionA7.Thenextsectiondescribesthedatacollec-tionandempiricalstrategyfortestingthesetheoriesofdemocracyinthestates.15Duringtheperiodunderstudyinthispaper,LatinoandAsianAmericanpopulationproportionsincreasedinmoststates.Further-more,theBlackpopulationofSouthernstatesincreasedaspartofthe“reverse”GreatMigrationsince1975.LaboratoriesofDemocraticBacksliding975https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
EmpiricallyTestingTheoriesofDemocracyToempiricallytestthesetheories,Icollecttime-seriesmeasuresofpoliticalcompetitiveness,polarization,partycontrol,anddemographicchange.IusedataonlegislativeseatsharesfromKlarner(2013)tomeasurelegislativecompetitiveness.Specifically,Icalculatestates’lowerlegislativechambercompeti-tivenessas−∣0:5−Dlower∣,whereDloweristhetwo-partyshareoflowerchamberseatsheldbyDemocrats,andupperchambercompetitivenessas−∣0:5−Dupper∣,whereDupperisthetwo-partyshareofupperchamberseatsheldbyDemocrats.16InrobustnesschecksintheAppendix,IuseanadditionalmeasureofelectoralratherthanlegislativecompetitivenessfromO’Brian(2019),whichIcodeas−∣0:5−Dvotes∣,whereDvotesisthetwo-partyshareofvotesinthestate’sU.S.Houseelection(s)thatwenttoDemocraticcandidates.17Asiscustomary,thesemeasuresaresmoothedintoroll-ingaveragesacrossthreeelectioncycles(e.g.,Ranney1976;ShufeldtandFlavin2012),butIlagtheminstatisticalmodelssuchthattheycaptureelectoralcompetitioninthethreepreviouselectioncyclespriortothestate’sdemocraticperformanceinyeart.LegislativepolarizationmeasuresarefromShorandMcCarty(2011).Iusetheaveragedistanceintheparties’legislativechambermedianswithineachstate.18Measuresofcompetitivenessandpolarizationarestandardizedtohaveameanof0andastandarddeviation1forclarity.Republicancontrolisabinaryvariablethattakesavalueof1ifthestateisunderunifiedRepublicancontroland0ifthestateisunderDemocraticordividedcontrol.19Stateracialdemo-graphicsarefromtheU.S.CensusBureau’s“bridged”1990–2019stateracepopulationestimates.20Imeasuredemographicchangeinfour-yearrollingaverages,buttheresultsarerobusttotheuseofdifferentyearincrements.StatepartycontroldataarefromKlarner(2013),whichIextendthrough2018usingNationalConferenceofStateLegislaturesdata.21IexcludeNebraskafromtheanalysesduetoitsnonpartisanunicamerallegislature.Itesttheoreticalpredictionswithadifference-in-differencesdesignthatexploitswithin-statevariation.Althoughthetruecausalmodelbetweencompetition,polarization,demographicchange,partycontrol,anddemocraticperformanceislikelytoinvolveastructureofhighlycomplexfeedbackrelationships,thisdesigneliminatestime-invariantdifferencesbetweenstates—themainpotentialsourceofbiasinestimatingtherelationshipbetweenmyinputmeasuresanddemo-craticperformance.22Isupplementtraditionaltwo-wayfixedeffectsmodelswithageneralizedsyntheticcontrolestimatorfrom(Xu2017)andalternativemethodsofaggregatingtreatmenteffectsfromCall-awayandSant’Anna(2020)thatavoidpotentialweightingproblemsinmultiperioddifference-in-differencesdesigns.RESULTSIpresentthemainresultsinTable1.TheresultsofModels1through3showthat,ontheirown,thereisamodestandsometimesstatisticallysignificantpositiverelationshipbetweencompetitionanddemocracyandnorelationshipbetweenpolarizationanddemocracy—butthereisalargenegativerelationshipbetweenRepublicancontrolanddemocracyinthestates.Acrossthemodelspecifications,theestimatesoftheeffectofRepublicancontrolofgovernmentarebetween0.442and0.481standarddeviationsofdemocraticperfor-mance,asubstantialamount.Theeffectofcompetition,incontrast,isbetween0.141and0.206standarddevi-ations,andtheeffectofpolarizationisverysmallandintheunexpectedlypositivedirection.Iamalsointerestedintheinteractionsofcompeti-tion,polarization,andRepublicancontrol.Polarizedparties(ortheRepublicanParty)mightonlyhaveanincentivetorestrictdemocracyincompetitivepoliticalenvironments.However,theresultsinTable1suggestthattheseinteractionsdolittletoexplaindynamicsinstatedemocracy.Theinteractionofcompetitionandpolarizationismodestlypositive,asistheinteractionofcompetitionandRepublicancontrol—bothcontrarytoexpectations(thoughalloftheinteractioncoefficientsarestatisticallyinsignificant).Duetorecentconcernabouttheweightingoftreat-mentestimatesinmultiperioddifference-in-differencesanalysisusingtwo-wayfixedeffects(Goodman-Bacon2018),Iusealternativeaggregationprocedurestoesti-matetheaveragetreatmenteffectonthetreated(ATT)ofRepublicancontrol.23InPanel(a)ofFigure5,IplottheresultsfromthreedifferenttypesofATTaggregationfromCallawayandSant’Anna(2020):dynamic,group,andsimple(group-time).Inadditiontousingdifferentaggregationprocedures,themodeldropsstatesthatwere“treated”(i.e.,underRepublicancontrol)inthefirstperiod,theyear16AlthoughImightideallywishforameasureofcompetitioninbothstatelegislaturesandexecutivebranches,moststudiesuselegislativemajoritysizeasthemainmeasureofcompetition,whetherintheU.S.Congress(Lee2009)orstatepolitics(Teele2018b)17O’Brian(2019)collectedvote-sharedatafromDavidandClaggett(2008)andCQPress’sVotingandElectionCollection.18Thechoiceofusingseparatevariablesforupperorlowerlegisla-tivechamberpolarization,ortheiraverage,doesnotaffecttheresults.19Futureresearchcandisaggregatepartisancontrolofeachlegisla-tivechamber,thegovernorship,andtheirinteractionstostudymoregranulareffectsofpartisancontrol.20Available:https://wonder.cdc.gov/bridged-race-v2019.html.21Available:https://www.ncsl.org/research/about-state-legislatures/partisan-composition.aspx.22Themaintwo-wayfixedeffectsmodelstakethefollowingformforstateiinyeart,inwhichXisatreatmentvariable,αiarestatefixedeffects,andδtareyearfixedeffects:yit=αiþδtþβXitþeit.23Specifically,two-wayfixedeffectsspecificationsareaweightedaverageofallpossibletwo-perioddifference-in-differencesestima-tors,whichisvulnerabletobiasiftreatmenteffectsvaryacrosstimeinmultiperioddesigns.JacobM.Grumbach976https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press
2000.24InPanel(b),IplottheeffectsofGOPcontrolusingthegeneralizedsyntheticcontrol(GSC)methodfromXu(2017).TheGSCtechniquerelaxestheparal-leltrendsassumptioninthedifference-in-differencesdesignsusedthroughoutthisarticlebycreatingsyn-theticcontrolunitsthatareweightedaveragesofthe“real”controlunits,eachconstructedtocloselymatchthepretreatmentdemocraticperformanceinstatesthatwilleventuallybetreatedbyGOPcontrol(forotherexamplesofGSCinpoliticalscience,seeGilens,Pat-terson,andHaines2021;Marbleetal.2021).ComparedwiththemainstateandyearfixedeffectsresultsinTable1,theresultsinFigure5showanevenlargereffectofRepublicancontrol.TheresultsinPanel(a)usingtheCallawayandSant’Anna(2020)estima-torsincreasemyconfidencethattheRepublicancon-trolfindingsarenotbeingdrivenbytheparticulartimingoftreatment(i.e.,changeinpartycontrol)andthetimeheterogeneityoftreatmenteffects,whereastheGSCestimatesinpanel(b)increasemyconfidencethattheeffectisrobusttoequalizingpretrendsindemocraticperformance.25IntheAppendix,Ishowthattheseresultsarerobustunderawidevarietyofconditions.First,Ireplicatetheseresultsusingtheadditivedemocracyindexdescribedearlierinwhicheachdemocracyindicatorisweightedequally.TheresultsinAppendixTableA2aresubstantivelyunchanged.Second,Ireplicatethemainanalysesusingameasureofpartisanelectoralcompetitiveness(i.e.,theclosenessofelections)ratherthanlegislativecompetitiveness(i.e,thenarrownessofpartisanlegislativemajorities).TableA3intheAppen-dixshowsresultsconsistentwiththemainresults,butwithoneimportantdifference.Althoughtheeffectsofcompetitiveness,polarization,andRepublicancontrolremainverysimilartothemainresults,theinteractionofcompetitivenessandRepublicancontrolisnegative,significant,andrelativelysubstantialinmagnitude(-0.262standarddeviationsofStateDemocracyIndexscores).AmongRepublican-controlledstates,inotherwords,thosewhoserecentelectionshavebeenespe-ciallycompetitivearethestatesthattakestepstoreducetheirdemocraticperformance.26InAppendixSectionA6,Ireplicatethemainana-lyseswithalternativemeasuresofdemocracy.Thefirstmeasurecoversliberalandelectoraldemocracy(using61totalindicators),andthesecondcoversliberal,electoral,andegalitariandemocracy(using116totalindicators).Theadditionalliberaldemocracyindica-torsextendthemeasure’scoveragetoissuesofcivillibertiesandfreedomfromstateauthorityinareassuchaspolicing,incarceration,andfreedomofthepress.TABLE1.ExplainingDynamicsinState-LevelDemocracyOutcome:StateDemocracyScoreModel1Model2Model3Model4Model5Model6Model7Competition0.2000.1700.1940.1690.134(0.107)(0.099)(0.099)(0.106)(0.114)Polarization0.0170.0240.0370.0270.028(0.131)(0.119)(0.111)(0.126)(0.121)Republican−0.462**−0.444**−0.435**−0.443**−0.475**(0.162)(0.159)(0.162)(0.154)(0.183)CompetitionPolarization0.082(0.066)PolarizationRepublican−0.013(0.198)CompetitionRepublican0.110(0.206)Constant−0.707***−0.683***−0.532***−0.535***−0.544***−0.533***−0.532***(0.068)(0.116)(0.093)(0.134)(0.136)(0.139)(0.135)StateFEsYesYesYesYesYesYesYesYearFEsYesYesYesYesYesYesYesN833833833833833833833R20.6830.6760.6990.7040.7060.7040.705Adj.R20.6560.6480.6730.6790.6800.6780.679Note:*p<0.05,**p<0.01,***p<0.001.24IntheCallawayandSant’Anna(2020)setup,treatmentcannotswitchbackoffonceitison.Inturn,IassignastatethatswitchestoRepublicancontrolanewunitfixedeffectonceitswitchesbacktodivided(orDemocratic)control.Theresultsarerobusttoexcludingthesestate-years.25ThespecificationinPanel(b)ofFigure5usesabivariate“two-way”specificationandsevenpretreatmentperiodstocreatesyntheticcontrolunits.AppendixFigureA3presentstheresultsofadditionalspecificationsthatvarythenumberofpretreatmentperiods.26InFigureA5intheAppendix,Isimulate100,000democracymeasureswithrandomweightsforeachindicator,andthenrunthemainregressionspecificationstoobtainadistributionofcoefficientsacrossmanyhypotheticalmeasures(analogoustotheBayesianbootstrap).TheRepublicaneffectremainsremarkablyrobust.LaboratoriesofDemocraticBacksliding977https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press Theegalitariandemocracyindicatorsincludemeasuresofeconomicinequality,women’srights,campaignfinancepolicy,laborrights,andLGBTrights,whichscholarshavearguedareintegraltotherealizationofdemocracyinpractice.Theresultsfromtheadditionalanalysesaresubstantivelyverysimilartotheanalysesusingthemain(electoral)StateDemocracyIndexmeasure,withRepublicancontrolsignificantlyreduc-ingdemocraticperformance,andlittleexplanatoryroleforotherpotentialcausesofdemocraticchange.RacialDemographicChangeandStateDemocracyInthissection,Iturntotheanalysisofracialdemo-graphicchangeanditsinteractionwithcompetition,polarization,andRepublicangovernance.Ifirstassessdescriptivetrends.Figure6plotsBlackandLatinopopulationchangeinthefivestatesthatexperiencethegreatestdemocraticbackslidingovertheperiod:Alabama,Ohio,NorthCarolina,Tennessee,andWis-consin.Thesestatestendtohaveabove-averageBlackpopulationshares,buttheyseelittlechangeovertime.Incontrast,thesestateshaverelativelylowLatinopopulationshares.TheirLatinopopulationsgrowgrad-uallyoverthisperiod.However,thisamountofgrowthisnotoutoftheordinary;thetrendsinthesestatescloselytracknationalaverages.Thisdescriptiveanaly-sisprovideslittleevidencethatlocalBlackorLatinopopulationchangemattersmuchforstatedemocraticperformance.Table2teststheoriesofdemographicthreatwiththemaindifference-in-differencesdesign.Theresultsareconsistentwiththedescriptiveanalysis:trendsinracialpopulationproportionshavelittleeffectonstatedemocraticperformance.Furthermore,althoughRepublicancontrolstillhasalargenegativeeffectondemocraticperformance,theinteractionofRepublicancontrolanddemographicchangegenerallymatterslittle.Unexpectedly,theonestatisticallysignificantcoefficientinvolvingdemographicchangeistheposi-tivecoefficientfortheinteractionofRepublicancon-trolandLatinopopulationchange,meaningthatRepublicanstateswithgreaterLatinopopulationgrowthreducedemocraticperformanceslightlylessthandootherRepublicanstates(thoughwithacoeffi-cientof0.325correspondingtoa1percentage-pointincreaseinstatepercentageLatino,ornearlytwostandarddeviations,thiseffectissmallinsubstantivemagnitude).27Thesefindingssuggestthatracialpoliticswithinstatesarenotcentraltodynamicsinstatedemocracy.28Thisdoesnotmeanthatraceisperipheraltodynamicsinstatedemocracy.Onthecontrary,theyareconsistentwithacentralroleofraceinnationalpoliticalconflict,especiallyatthemasslevel(ParkerandBarreto2014;Sides,Tesler,andVavreck2018).Anumberofimpor-tantstudiesshowevidenceofracialthreatandcontes-tationathighlylocalizedlevels(e.g.,Enos2017).ButinaneraofhighlynationalizedAmericanpolitics(HopkinsFIGURE5.EffectofRepublicanControlonDemocraticPerformance(a) Republican Control Effect UsingCallaway and Sant’ Anna Estimator−2−101DynamicGroupGroup−TimeATT AggregationEffect of GOP Control onDemocratic Performance(b) Republican Control Effect Using Synthetic Control−3−2−10−10−505Years Relative to GOP ControlDemocratic PerformanceNote:Panel(a)showsresultsusingtheCallawayandSant’AnnaestimatoralternativeATTaggregationmethods.Panel(b)showstheresultsofageneralizedsyntheticcontrolanalysis.27Ininterpretingthisresult,itisimportanttoconsidertheconsider-ablepoliticalheterogeneityofLatinoAmericansanditsrelationshiptogeographyandnationalorigingroup(delaGarzaetal.2019).28Thispaper’sfocusonwithin-statechangeisalsothereasonitsfindingsaboutracialdemographicsdifferfromthoseofBiggersandHanmer(2017),whofindthattheinteractionofRepublicancontrolwithpercentBlackorLatinoisassociatedwiththeimplementationofvoterIDlaws.Thispaper’sdifference-in-differencesdesignsuggeststhatchangeindemographicsisnotarelevantfactor,whetheronitsownorinteractedwithRepublicancontrol.JacobM.Grumbach978https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press FIGURE6.BlackandLatinoPopulationChangeintheStatesALNCOHTNWIALNCOHTNWIBlackLatino20002005201020152000200520102015010203040YearPercent of State PopulationTABLE2.RacialDemographicChangeandStateDemocracyOutcome:StateDemocracyScoreModel1Model2Model3Model4Δ%Black−0.012−0.1050.0580.071(0.249)(0.266)(0.374)(0.253)Δ%Latino−0.0190.020−0.010−0.174(0.202)(0.189)(0.207)(0.186)Competition0.317(0.165)Polarization0.007(0.199)Republican−0.726**(0.252)Δ%BlackCompetition0.014(0.280)Δ%LatinoCompetition−0.140(0.095)Δ%BlackPolarization0.094(0.226)Δ%LatinoPolarization−0.029(0.130)Δ%BlackRepublican−0.140(0.280)Δ%LatinoRepublican−0.325*(0.156)Constant−0.673***−0.670***−0.694***−0.358*(0.166)(0.166)(0.169)(0.177)StateFEsYesYesYesYesYearFEsYesYesYesYesN833833833833R20.6760.6850.6760.705Adj.R20.6480.6570.6470.678Note:*p<0.05,**p<0.01,***p<0.001.LaboratoriesofDemocraticBacksliding979https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press 2018),whenitcomestostategovernmentalchoicesoverdemocraticinstitutions,thecriticalquestionisnotaboutracialpoliticswithinastatebutwhetherthestategov-ernmentispartofthenationalRepublicanParty.Therefore,thesefindings,suggestacontrastfromtheracialpoliticsofJimCrow.Althoughcontemporarystateelectorallegislation,likethatoftheJimCrowera,hasbeenfoundbycourtstohavebeen“motivatedatleastinpartbyanunconstitutionalintenttotargetAfricanAmericanvoters,”29battlesovervotingrightsfromthe1890sthrough1970sprimarilyinvolvedbattlesbetweenlargelandowners,Blackactivists,andother“indigenous”pro-andantidemocracyinterestinSouth-ernstates(Mickey2015).Insuchapoliticaleconomy,states’racialdemographicsplayacentralroleinexplainingvariationinsubnationaldemocratization.Incontrast,theresultsinthisarticleemphasizetheimportanceofnationalpoliticalforces.CONCLUSIONDespitethenationalfocusofmuchcontemporarydis-courseaboutdemocraticbackslidingintheUSandabroad,stategovernmentshaveconstitutionalauthor-itytostructureandadministermanyofthemostimpor-tantdemocraticinstitutionsintheAmericanpoliticalsystem.Thisarticlecreatesanewmeasureofelectoraldemocracyinthe50statesfrom2000to2018,basedon51indicators.IntheAppendix,Iconstructadditionalmeasuresthatalsocoverliberaldemocracyandegali-tariandemocracy.Themeasure,theStateDemocracyIndex,suggeststhattherehavebeendramaticshiftsindemocraticperformanceintheAmericanstatesoverthisperiod.Insomestates,democracyexpandedininclusiveways,expandingaccesstopoliticalparticipation,reducingtheauthoritarianuseofpolicepowers,andmakingelectoralinstitutionsmorefair.Inotherstates,how-ever,democracynarroweddramatically,asstategov-ernmentsgerrymandereddistrictsandcreatednewbarrierstoparticipationandrestrictionsonthefran-chise.Mymeasureopensupnewopportunitiesforresearchonquestionsrelatedtorepresentationanddemocracy,aswellasfederalismandstateandlocalpolitics.Scholarsmightbeinterestedininvestigatingtheroleofinterestgroupsormoneyinpoliticsonstatedemo-craticperformance(AnziaandMoe2017;Hertel-Fernandez2016),perhapsbyexploitingvariationinstatecampaignfinancepolicy(Barber2016;LaRajaandSchaffner2015)orelectiontiming(Anzia2011).Othersmightstudyhowstatedemocracyisaffectedbydecliningstateandlocalpoliticsjournalism(Moskowitz2021)orbyvoters’attitudestowarddemocraticinsti-tutions(GrahamandSvolik2020;MillerandDavis2020;Welzel2007).Thereisespeciallygreatpotentialforbehavioralscholarsofraceandethnicpoliticstoinvestigatetherelationshipbetweenracialattitudes,attitudestowarddemocracy,andstatedemocraticper-formance(e.g.,Jefferson2021;Mutz2018;WeaverandProwse2020).Likecomparativeandpoliticaleconomyscholarshiponwhether“democracycausesgrowth”(Acemogluetal.2019),scholarscanalsousetheStateDemocracyIndexasanexplanatoryvariabletostudytheeffectofdemocraticperformanceoneconomicperformance,socioeconomicoutcomesamongresi-dents,andpublicattitudessuchastrust.Comparativescholarscanusemymeasurementstrategytocreatenewmeasuresofdemocraticperformanceinsubna-tionalunitsinoneormoreothercountries,potentiallyconstructingcomprehensivecross-nationalmeasuresofsubunitdemocracyinpoliticalfederations.Inthisarticle,IusetheStateDemocracyIndextotestasetofprominenttheoriesofthecausesofdemocraticexpansionandbackslidingintheUS.DrawingonAmericanandcomparativedemocracyliteratures,Ideveloppredictionsaboutthedriversofdemocraticexpansionandbacksliding.Iestimatetheeffectsofpoliticalcompetition,polarization,andracialdemo-graphicchangeonstates’democraticperformance.Theresultssuggestthatnoneofthesefactorsiscentraltodynamicsinstatedemocraticperformance.Repub-licancontrolofstategovernment,however,consis-tentlyandprofoundlyreducesstatedemocraticperformanceduringthisperiod.ThelargeeffectsofRepublicancontrol,contrastedwiththeminimaleffectsofwithin-statedynamics,addressthenationalizationofAmericanpoliticsinrecentdecades.PoliticalinvestmentsbygroupsintheDemocraticandRepublicanpartycoalitionshavemadethepartycoalitionsmorenationallycoordinated(Grumbach2019;HackerandPierson2020;Hertel-Fernandez2019).Votersareincreasinglyfocusedonnationalratherthanstateandlocalpolitics(Hopkins2018),inpartduetothedeclineofstateandlocalpoliticsjournalism(MartinandMcCrain2019;Moskowitz2021).Thistransformationmeansthatregardlessoftheparticularcircumstancesorgeogra-phy,stategovernmentscontrolledbysamepartybehavesimilarlywhentheytakepower.TheRepublican-controlledgovernmentsofstatesasdistinctasAlabama,Wisconsin,Ohio,andNorthCarolinahavetakensimilaractionswithrespecttodemocraticinsti-tutions.Moreresearchisneededtolinkthisissueofstate-leveldemocraticperformanceintheUStomicro-levelbehavioralresearchontherelationshipbetweensocialcleavages,theRepublicanParty,andsupportfordemocracy.Thefindingsinthisarticleareconsistentwithanimportantrolefornational(butnotstate-level)racialthreat(e.g.,Mutz2018;ParkerandBarreto2014).Bartels(2020,22752),forinstance,findsthat“substantialnumbersofRepublicansendorsestate-mentscontemplatingviolationsofkeydemocraticnorms,includingrespectforthelawandfortheout-comesofelectionsandeschewingtheuseofforceinpursuitofpoliticalends”andthat“[t]hestrongestpredictorbyfaroftheseantidemocraticattitudesisethnicantagonism—especiallyconcernsaboutthe29https://www.nbcnews.com/politics/elections/north-carolina-court-blocks-voter-id-law-discriminatory-intent-n1279474.JacobM.Grumbach980https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press politicalpowerandclaimsongovernmentresourcesofimmigrants,African-Americans,andLatinos.”How-ever,leftunexploredinthisarticleistheroleofotherimportantsocialcleavages,includingthosebasedongender,religion,andsexuality.Incontrasttomymeasures,cross-nationalmeasuresofdemocracysometimescovermuchlongerstretchesoftime.V-Dem,forinstance,measuresdemocraticperformanceforcountriesasfarbackastheyear1789—thoughthisisnotwithoutitschallenges(forexample,duringperiodsofrapidchangestoU.S.democracy,suchasduringReconstruction).Still,itisaworthygoaltoconstructaStateDemocracyIndexthatcoversthetransformationalchangestothefran-chise,civilliberties,andothercomponentsofdemoc-racythatoccurredinearlierperiodsofU.S.history.Keyssar(2000)andothershaveengagedinthiskindofhistoricalanalysisofchangesinvotingrights.Perhapsmoreimportantly,alongertimeframewouldcontextualizethemagnitudeofrecentshiftsinstate-leveldemocracy.Thisarticleprovidesclearevi-denceofimportantchangesindemocraticperfor-mance,suchastherapiddeclineofdemocracyinstatessuchasNorthCarolinasince2010.Buttheserecentchangeshaveoccurredonanarrowerrangeofthedemocracydimensionthanhavethoseinearlierperiods,when,forexample,statesdifferedintermsofthelegalityofslaveryandthefranchiseforwomen.Despitesometroublingexamplesinstate-leveldemocracyinrecentyears,theydonotcomeclosetotheprofounddifferencesinregimetypethatexistedbetweenstatesintheerasbeforethetwentieth-centurycivilrightsperiod.Atthesametime,amoresignificantdemocraticcollapseislikelytobepresagedbythekindsofdemocraticbackslidingdescribedinthisarticle—whichcanentrenchminorityrule,curtaildissent,andlimitparticipationindemocraticinstitu-tions.Thisstudycombineswhatareattimesdisparatedis-cussionsofAmericandemocracy.Idrawuponscholar-shipondemocraticexpansionandbackslidingintheUSandothernation-states,whilealsosynthesizingmanydistinctinquiriesintostate-levelactioninelectionadmin-istration,gerrymandering,andobserveddemocraticout-comes.Intheuseofadeepwellofdata,Ihopethatthisstudycontributestoquantitativemeasurementandthe-orytestingoflarge-scale,substantivelyprofoundques-tionsinpoliticalscienceandpoliticaleconomy.SUPPLEMENTARYMATERIALSToviewsupplementarymaterialforthisarticle,pleasevisithttp://doi.org/10.1017/S0003055422000934.DATAAVAILABILITYSTATEMENTResearchdocumentationanddatathatsupportthefindingsofthisstudyareopenlyavailableattheAmericanPoliticalScienceReviewDataverse:https://doi.org/10.7910/DVN/JNV3XO.ACKNOWLEDGMENTSForhelpfulfeedbackanddiscussions,theauthorthanksZackBeauchamp,AdamBonica,RuthCollier,MicheleEpstein,JacobHacker,CharlotteHill,HakeemJefferson,JamilaMichener,RobMickey,PaulPierson,PhilRocco,andEricSchickler,aswellasseminarparticipantsatAmericanUniversity,JohnsHopkins,StanfordCDDRL,UCBerkeleyRWAP,UCLA,UCSD,UniversityofMinnesota,UniversityofOregon,UWCSSS,UWWISIR,Vanderbilt,theAmericanDemocracyCollaborative,andtheDemoc-racyintheStatesMeetingatPrincetonCSDP.IthankChrisWarshaw,NickStephanopoulos,andtheMITElectionLabforgraciouslysharingdataandSteveLevitskyandDanZiblattforinspiringthetitleofthispaper.CONFLICTOFINTERESTTheauthordeclaresnoethicalissuesorconflictsofinterestinthisresearch.ETHICALSTANDARDSTheauthoraffirmsthisresearchdidnotinvolvehumansubjects.REFERENCESAbrajano,Marisa,andZoltanL.Hajnal.2017.WhiteBacklash:Immigration,Race,andAmericanPolitics.Princeton,NJ:PrincetonUniversityPress.Acemoglu,Daron,SureshNaidu,PascualRestrepo,andJamesA.Robinson.2019.“DemocracyDoesCauseGrowth.”JournalofPoliticalEconomy127(1):47–100.Aneja,AbhayP.,andCarlosF.Avenancio-León.2019.“DisenfranchisementandEconomicInequality:DownstreamEffectsofShelbyCountyv.Holder.”AEAPapersandProceedings109(May):161–65.Angrist,JoshuaD.,andJörn-SteffenPischke.2008.MostlyHarmlessEconometrics:AnEmpiricist’sCompanion.Princeton,NJ:PrincetonUniversityPress.Anzia,SarahF.2011.“ElectionTimingandtheElectoralInfluenceofInterestGroups.”TheJournalofPolitics73(2):412–27.Anzia,SarahF.,andTerryM.Moe.2017.“DoPoliticiansUsePolicytoMakePolitics?TheCaseofPublic-SectorLaborLaws.”AmericanPoliticalScienceReview110(4):763–77.Barber,Michael.2016.“IdeologicalDonors,ContributionLimits,andthePolarizationofStateLegislatures.”TheJournalofPolitics78(1):296–310.Bartels,LarryM.2020.“EthnicAntagonismErodesRepublicans’CommitmenttoDemocracy.”ProceedingsoftheNationalAcademyofSciences117(37):22752–759.Benton,AllysonLucinda.2012.“Bottom-UpChallengestoNationalDemocracy:Mexico’s(Legal)SubnationalAuthoritarianEnclaves.”ComparativePolitics44(3):253–71.Biggers,DanielR.,andMichaelJ.Hanmer.2017.“UnderstandingtheAdoptionofVoterIdentificationLawsintheAmericanStates.”AmericanPoliticsResearch45(4):560–88.Bobo,Lawrence,andVincentL.Hutchings.1996.“PerceptionsofRacialGroupCompetition:ExtendingBlumer’sTheoryofGroupPositiontoaMultiracialSocialContext.”AmericanSociologicalReview61(6):951–72.LaboratoriesofDemocraticBacksliding981https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press Brettschneider,Corey.2010.DemocraticRights:TheSubstanceofSelf-Government.Princeton,NJ:PrincetonUniversityPress.Burden,BarryC.,DavidT.Canon,KennethR.Mayer,andDonaldP.Moynihan.2014.“ElectionLaws,Mobilization,andTurnout:TheUnanticipatedConsequencesofElectionReform.”AmericanJournalofPoliticalScience58(1):95–109.Callaway,Brantly,andPedroH.C.Sant’Anna.2020.“Difference-in-DifferenceswithMultipleTimePeriods.”JournalofEconometrics225(2):200–30.Caughey,Devin,andChristopherWarshaw.2016.“TheDynamicsofStatePolicyLiberalism,1936–2012.”AmericanJournalofPoliticalScience60(4):899–913.Caughey,Devin,andChristopherWarshaw.2018.“PolicyPreferencesandPolicyChange:DynamicResponsivenessintheAmericanStates,1936–2014.”AmericanPoliticalScienceReview112(2):249–66.Caughey,Devin,andEricSchickler.2016.“SubstanceandChangeinCongressionalIdeology:NOMINATEandItsAlternatives.”StudiesinAmericanPoliticalDevelopment30(October):128–46.Collier,David,andStevenLevitsky.1997.“DemocracywithAdjectives:ConceptualInnovationinComparativeResearch.”WorldPolitics49(3):430–51.Dahl,RobertA.1989.DemocracyandItsCritics.NewHaven,CT:YaleUniversityPress.Dahl,RobertA.2003.HowDemocraticistheAmericanConstitution?NewHaven,CT:YaleUniversityPress.David,PaulT.,andWilliamClaggett.2008.PartyStrengthintheUnitedStates:1872-1996[computerfile].AnnArbor:InstituteforSocialResearchUniversityofMichigan[producer].Inter-UniversityConsortiumforPoliticalandSocialResearch[distributor].delaGarza,RodolfoO.,LouisDeSipio,AngeloFalcon,F.ChrisGarcia,andJohnA.García.2019.LatinoVoices:Mexican,PuertoRican,andCubanPerspectivesonAmericanPolitics.NewYork:Routledge.Dionne,E.J.,Jr.,NormanJ.Ornstein,andThomasE.Mann.2017.OneNationafterTrump:AGuideforthePerplexed,theDisillusioned,theDesperate,andtheNot-YetDeported.NewYork:St.Martin’sPress.DuBois,W.E.B.1935.BlackReconstructioninAmerica:AnEssaytowardaHistoryofthePartWhichBlackFolkPlayedintheAttempttoReconstructDemocracyinAmerica,1860—1880.NewYork:HarcourtBrace.Enos,RyanD.2017.TheSpacebetweenUs:SocialGeographyandPolitics.NewYork:CambridgeUniversityPress.Erikson,RobertS.,GeraldC.Wright,andJohnP.Mclver.1993.StatehouseDemocracy:PublicOpinionandPolicyintheAmericanStates.NewYork:CambridgeUniversityPress.Estlund,David.2009.DemocraticAuthority:APhilosophicalFramework.Princeton,NJ:PrincetonUniversityPress.Flavin,Patrick,andWilliamW.Franko.2017.“Government’sUnequalAttentivenesstoCitizens’PoliticalPriorities.”PolicyStudiesJournal45(4):659–87.Foner,Eric.1988.Reconstruction:America’sUnfinishedRevolution,1863-1877.NewYork:Harper&Row.Garand,JamesC.1988.“ExplainingGovernmentGrowthintheUSStates.”AmericanPoliticalScienceReview82(3):837–49.Gay,Claudine.2007.“LegislatingwithoutConstraints:TheEffectofMinorityDistrictingonLegislators’ResponsivenesstoConstituencyPreferences.”TheJournalofPolitics69(2):442–56.Gerber,AlanS.,GregoryA.Huber,andSethJ.Hill.2013.“IdentifyingtheEffectofAll-MailElectionsonTurnout:StaggeredReformintheEvergreenState.”PoliticalScienceResearchandMethods1(1):91–116.Gessen,Masha.2016.“Autocracy:RulesforSurvival.”NewYorkReview,November10.https://www.nybooks.com/daily/2016/11/10/trump-election-autocracy-rules-for-survival/.Gibson,EdwardL.2005.“BoundaryControl:SubnationalAuthoritarianisminDemocraticCountries.”WorldPolitics58(1):101–32.Gibson,EdwardL.2013.BoundaryControl:SubnationalAuthoritarianisminFederalDemocracies.NewYork:CambridgeUniversityPress.Gilens,Martin,ShawnPatterson,andPavielleHaines.2021.“CampaignFinanceRegulationsandPublicPolicy.”AmericanPoliticalScienceReview115(3):1074–81.Gilley,Bruce.2010.“DemocraticEnclavesinAuthoritarianRegimes.”Democratization17(3):389–415.Giraudy,Agustina.2015.DemocratsandAutocrats:PathwaysofSubnationalUndemocraticRegimeContinuitywithinDemocraticCountries.Oxford:OxfordUniversityPress.Glaude,EddieS.,Jr.2017.DemocracyinBlack:HowRaceStillEnslavestheAmericanSoul.NewYork:BroadwayBooks.Gleditsch,KristianS.,andMichaelD.Ward.1997.“DoubleTake:AReexaminationofDemocracyandAutocracyinModernPolities.”JournalofConflictResolution41(3):361–83.Goodman-Bacon,Andrew.2018.“Difference-in-DifferenceswithVariationinTreatmentTiming.”NationalBureauofEconomicResearchWorkingPaper25018.Graham,MatthewH.,andMilanW.Svolik.2020.“DemocracyinAmerica?Partisanship,Polarization,andtheRobustnessofSupportforDemocracyintheUnitedStates.”AmericanPoliticalScienceReview114(2):392–409.Grumbach,JacobM.2018.“FromBackwaterstoMajorPolicymakers:PolicyPolarizationintheStates,1970-2014.”PerspectivesonPolitics16(2):416–35.Grumbach,JacobM.2019.“InterestGroupActivistsandthePolarizationofStateLegislatures.”LegislativeStudiesQuarterly45(1):5–34.Grumbach,JacobM.2022.“ReplicationDatafor:LaboratoriesofDemocraticBacksliding.”HarvardDataverse.Dataset.https://doi.org/10.7910/DVN/JNV3XO.Grumbach,JacobM.,andCharlotteHill.2021.“RocktheRegistration:SameDayRegistrationIncreasesTurnoutofYoungVoters.”TheJournalofPolitics84(1):405–17.Gulasekaram,Pratheepan,andS.KarthickRamakrishnan.2015.TheNewImmigrationFederalism.NewYork:CambridgeUniversityPress.Hacker,JacobS.,andPaulPierson.2005.OffCenter:TheRepublicanRevolutionandtheErosionofAmericanDemocracy.NewHaven,CT:YaleUniversityPress.Hacker,JacobS.,andPaulPierson.2010.“Winner-Take-AllPolitics:PublicPolicy,PoliticalOrganization,andthePrecipitousRiseofTopIncomesintheUnitedStates.”Politics&Society38(2):152–204.Hacker,JacobS.,andPaulPierson.2020.LetThemEatTweets:HowtheRightRulesinanAgeofExtremeInequality.NewYork:Liveright.Hajnal,ZoltanL.,andTaekuLee.2011.WhyAmericansDon’tJointheParty:Race,Immigration,andtheFailure(ofPoliticalParties)toEngagetheElectorate.Princeton,NJ:PrincetonUniversityPress.Harbers,Imke,JosBartman,andEnrikevanWingerden.2019.“ConceptualizingandMeasuringSubnationalDemocracyacrossIndianStates.”Democratization26(7):1154–75.Hero,RodneyE.,andCarolineJ.Tolbert.1996.“ARacial/EthnicDiversityInterpretationofPoliticsandPolicyintheStatesoftheUS.”AmericanJournalofPoliticalScience40(3):851–71.Hertel-Fernandez,Alexander.2014.“WhoPassesBusiness’s“ModelBills”?PolicyCapacityandCorporateInfluenceinUSStatePolitics.”PerspectivesonPolitics12(3):582–602.Hertel-Fernandez,Alexander.2016.“ExplainingLiberalPolicyWoesintheStates:TheRoleofDonors.”PS:PoliticalScience&Politics49(3):461–65.Hertel-Fernandez,Alexander.2019.StateCapture:HowConservativeActivists,BigBusinesses,andWealthyDonorsReshapedtheAmericanState—andtheNation.NewYork:OxfordUniversityPress.Hill,KimQuaile.1994.DemocracyintheFiftyStates.Lincoln:UniversityofNebraskaPress.Hollyer,JamesR.,B.PeterRosendorff,andJamesRaymondVreeland.2011.“DemocracyandTransparency.”TheJournalofPolitics73(4):1191–205.Hopkins,DanielJ.2018.TheIncreasinglyUnitedStates.Chicago,IL:UniversityofChicagoPress.Jefferson,Hakeem.2021.“StormingtheU.S.CapitolWasaboutMaintainingWhitePowerinAmerica.”FiveThirtyEight.January8.https://fivethirtyeight.com/features/storming-the-u-s-capitol-was-about-maintaining-white-power-in-america/.Jordan,MartyP.,andMattGrossmann.2016.TheCorrelatesofStatePolicyProjectv.1.5[computerfile].EastLansing:InstituteforJacobM.Grumbach982https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press PublicPolicyandSocialResearch[producer],MichiganStateUniversity[distributor].Kaufman,Aaron,GaryKing,andMayyaKomisarchik.2019.“HowtoMeasureLegislativeDistrictCompactnessIfYouOnlyKnowItWhenYouSeeIt.”AmericanJournalofPoliticalScience65(3):533–50.Keyssar,Alexander.2000.TheRighttoVote:TheContestedHistoryofDemocracyintheUnitedStates.NewYork:BasicBooks.King,Desmond.2017.“ForcefulFederalismagainstAmericanRacialInequality.”GovernmentandOpposition52(2):356–82.Klarner,Carl.2013.“StatePartisanBalanceData,1937–2011”,https://doi.org/10.7910/DVN/LZHMG3,HarvardDataverse,V1.Kousser,J.Morgan.1974.TheShapingofSouthernPolitics:SuffrageRestrictionandtheEstablishmentoftheOne-Partysouth,1880-1910.NewHaven,CT:YaleUniversityPress.LaRaja,RaymondJ.,andBrianF.Schaffner.2015.“CampaignFinanceandPoliticalPolarization:WhenPuristsPrevail.”AnnArbor:UniversityofMichiganPress.Lax,JeffreyR.,andJustinH.Phillips.2009.“GayRightsintheStates:PublicOpinionandPolicyResponsiveness.”AmericanPoliticalScienceReview103(3):367–86.Lax,JeffreyR.,andJustinH.Phillips.2012.“TheDemocraticDeficitintheStates.”AmericanJournalofPoliticalScience56(1):148–66.Lee,FrancesE.2009.BeyondIdeology:Politics,Principles,andPartisanshipintheUSSenate.Chicago,IL:UniversityofChicagoPress.Levitsky,Steven,andDanielZiblatt.2018.HowDemocraciesDie.NewYork:BroadwayBooks.Levitsky,Steven,andLucanA.Way.2010.CompetitiveAuthoritarianism:HybridRegimesaftertheColdWar.NewYork:CambridgeUniversityPress.Li,Quan,MichaelJ.Pomante,andScotSchraufnagel.2018.“CostofVotingintheAmericanStates.”ElectionLawJournal:Rules,Politics,andPolicy17(3):234–47.Lieberman,RobertC.,SuzanneMettler,ThomasB.Pepinsky,KennethM.Roberts,andRichardValelly.2019.“TheTrumpPresidencyandAmericanDemocracy:AHistoricalandComparativeAnalysis.”PerspectivesonPolitics17(2):470–79.Lindberg,StaffanI.,MichaelCoppedge,JohnGerring,andJanTeorell.2014.“V-Dem:ANewWaytoMeasureDemocracy.”JournalofDemocracy25(3):159–69.Lipset,SeymmyMartin.1960.PoliticalMan:TheSocialBasesofPolitics.GardenCity,NY:AnchorBooks.Marble,William,SalmaMousa,AlexandraA.Siegel,andAla’Alrababa’h.2021.“CanExposuretoCelebritiesReducePrejudice?TheEffectofMohamedSalahonIslamophobicBehaviorsandAttitudes.”AmericanPoliticalScienceReview115(4):1111–28.Mares,Isabela.2015.FromOpenSecretstoSecretVoting:DemocraticElectoralReformsandVoterAutonomy.NewYork:CambridgeUniversityPress.Martin,GregoryJ.,andJoshuaMcCrain.2019.“LocalNewsandNationalPolitics.”AmericanPoliticalScienceReview113(2):372–84.McCarty,Nolan,KeithT.Poole,andHowardRosenthal.2006.PolarizedAmerica:TheDanceofIdeologyandUnequalRiches,Vol.5.Cambridge,MA:MITPress.McGhee,Eric,CharlotteHill,andMindyRomero.2021.“TheRegistrationandTurnoutEffectsofAutomaticVoterRegistration.”SocialScienceResearchNetworkWorkingPaper3933442.https://ssrn.com/abstract=3933442.McMann,KellyM.2018.“MeasuringSubnationalDemocracy:TowardImprovedRegimeTypologiesandTheoriesofRegimeChange.”Democratization25(1):19–37.Meltzer,AllanH.,andScottF.Richard.1981.“ARationalTheoryoftheSizeofGovernment.”JournalofPoliticalEconomy89(5):914–27.Michener,Jamila.2018.FragmentedDemocracy:Medicaid,Federalism,andUnequalPolitics.NewYork:CambridgeUniversityPress.Mickey,Robert.2015.PathsoutofDixie:TheDemocratizationofAuthoritarianEnclavesinAmerica’sDeepSouth,1944-1972,Vol.147.Princeton,NJ:PrincetonUniversityPress.Miller,LisaL.2008.ThePerilsofFederalism:Race,Poverty,andthePoliticsofCrimeControl.Oxford:OxfordUniversityPress.Miller,WakkenE.,andDonaldE.Stokes.1963.“ConstituencyInfluenceinCongress.”AmericanPoliticalScienceReview57(1):45–56.Miller,StevenV.,andNicholasT.Davis.2020.“TheEffectofWhiteSocialPrejudiceonSupportforAmericanDemocracy.”TheJournalofRace,Ethnicity,andPolitics6(2):334–51.Mills,CharlesW.2017.BlackRights/WhiteWrongs:TheCritiqueofRacialLiberalism.NewYork:OxfordUniversityPress.Moskowitz,DanielJ.2021.“LocalNews,Information,andtheNationalizationofUSElections.”AmericanPoliticalScienceReview115(1):114–29.Mutz,DianaC.2018.“StatusThreat,NotEconomicHardship,Explainsthe2016PresidentialVote.”ProceedingsoftheNationalAcademyofSciences115(19):E4330–39.Myers,Dowell,andMorrisLevy.2018.“RacialPopulationProjectionsandReactionstoAlternativeNewsAccountsofGrowingDiversity.”TheANNALSoftheAmericanAcademyofPoliticalandSocialScience677(1):215–28.O’Brian,NeilA.2019.“One-PartyStatesandLegislatorExtremismintheUSHouse,1876–2012.”TheJournalofPolitics81(4):1223–39.Pacheco,Julianna.2013.“TheThermostaticModelofResponsivenessintheAmericanStates.”StatePolitics&PolicyQuarterly13(3):306–32.Parker,ChristopherS.,andMattA.Barreto.2014.ChangeTheyCan’tBelievein:TheTeaPartyandReactionaryPoliticsinAmerica-UpdatedEdition.Princeton,NJ:PrincetonUniversityPress.Phillips,Anne.1991.EngenderingDemocracy.UniversityPark:PennsylvaniaStateUniversityPress.Przeworski,Adam.1986.CapitalismandSocialDemocracy.NewYork:CambridgeUniversityPress.Quinn,KevinM.2004.“BayesianFactorAnalysisforMixedOrdinalandContinuousResponses.”PoliticalAnalysis12(4):338–53.Rakner,Lise,andNicolasVandeWalle.2009.“DemocratizationbyElections?OppositionWeaknessinAfrica.”JournalofDemocracy20(3):108–21.Ranney,Austin.1976.“PartiesinStatePolitics.”InPoliticsintheAmericanStates:AComparativeAnalysis,eds.HerbertJacobandKennethN.Vines,51–92.Boston,MA:Little,Brown.Rigby,Elizabeth,andGeraldC.Wright.2013.“PoliticalPartiesandRepresentationofthePoorintheAmericanstates.”AmericanJournalofPoliticalScience57(3):552–65.Riker,WilliamH.1964.Federalism:Origin,Operation,Significance.Boston,MA:Little,Brown.Riker,WilliamH.1975.“Federalism.”InHandbookofPoliticalScience:GovernmentalInstitutionsandProcesses,eds.FredI.GreensteinandNelsonW.Polsby,93–172.Reading,MA:Addison-Wesley.Rocco,Philip.2021.“LaboratoriesofWhat?AmericanFederalismandthePoliticsofDemocraticSubversion.”InDemocraticResilience:CantheUSWithstandRisingPolarization?eds.RobertLieberman,SuzanneMettler,andKenRoberts,297–319.NewYork:CambridgeUniversityPress.Rogers,Steven.2017.“ElectoralAccountabilityforStateLegislativeRollCallsandIdeologicalRepresentation.”AmericanPoliticalScienceReview111(3):555–71.Sartori,Giovanni.1970.“ConceptMisformationinComparativePolitics.”AmericanPoliticalScienceReview64(4):1033–53.Schickler,Eric.2016.RacialRealignment:TheTransformationofAmericanLiberalism,1932–1965.Princeton,NJ:PrincetonUniversityPress.Schuit,Sophie,andJonC.Rogowski.2017.“Race,Representation,andtheVotingRightsAct.”AmericanJournalofPoliticalScience61(3):513–26.Schumpeter,JosephA.1942.Capitalism,SocialismandDemocracy.NewYork:Harper&Brothers.Shapiro,Ian.2009.TheStateofDemocraticTheory.Princeton,NJ:PrincetonUniversityPress.Shelby,Tommie.2005.IWhoAreDark:ThePhilosophicalFoundationsofBlackSolidarity.Cambridge,MA:HarvardUniversityPress.Shor,Boris,andNolanMcCarty.2011.“TheIdeologicalMappingofAmericanLegislatures.”AmericanPoliticalScienceReview105(3):530–51.Shufeldt,Gregory,andPatrickFlavin.2012.“TwoDistinctConcepts:PartyCompetitioninGovernmentandElectoralCompetitionintheAmericanStates.”StatePolitics&PolicyQuarterly12(3):330–42.LaboratoriesofDemocraticBacksliding983https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press Sides,John,MichaelTesler,andLynnVavreck.2018.IdentityCrisis.Princeton,NJ:PrincetonUniversityPress.Simonovits,Gabor,AndrewM.Guess,andJonathanNagler.2019.“ResponsivenesswithoutRepresentation:EvidencefromMinimumWageLawsinUSStates.”AmericanJournalofPoliticalScience63(2):401–10.Soss,Joe,andVeslaWeaver.2017.“PoliceAreMyGovernment:Politics,PoliticalScience,andthePolicingofRace–ClassSubjugatedCommunities.”AnnualReviewofPoliticalScience20:565–91.Stephanopoulos,Nicholas,andChristopherWarshaw.2020.“TheImpactofPartisanGerrymanderingonPoliticalParties.”LegislativeStudiesQuarterly45(4):609–43.Stewart,Charles,III,StephenAnsolabehere,andNathanielPersily.2016.“RevisitingPublicOpiniononVoterIdentificationandVoterFraudinanEraofIncreasingPartisanPolarization.”StanfordLawReview68(June):1455–89.Teele,DawnLangan.2018a.ForgingtheFranchise:ThePoliticalOriginsoftheWomen’sVote.Princeton,NJ:PrincetonUniversityPress.Teele,DawnLangan.2018b.“HowtheWestWasWon:Competition,Mobilization,andWomen’sEnfranchisementintheUnitedStates.”TheJournalofPolitics80(2):442–61.Treier,Shawn,andSimonJackman.2008.“DemocracyasaLatentVariable.”AmericanJournalofPoliticalScience52(1):201–17.Weaver,VeslaM.,andGwenProwse.2020.“RacialAuthoritarianisminUSDemocracy.”Science369(6508):1176–78.Weiner,Myron.1965.“PoliticalIntegrationandPoliticalDevelopment.”TheAnnalsoftheAmericanAcademyofPoliticalandSocialScience358(1):52–64.Weissert,CarolS.2011.“BeyondMarbleCakesandPicketFences:WhatUSFederalismScholarsCanLearnfromComparativeWork.”TheJournalofPolitics73(4):965–79.Welzel,Christian.2007.“AreLevelsofDemocracyAffectedbyMassAttitudes?TestingAttainmentandSustainmentEffectsonDemocracy.”InternationalPoliticalScienceReview28(4):397–424.Wollheim,Richard.2016.“AParadoxintheTheoryofDemocracy.”InDemocracy:AReader,eds.RicardoBlaugandJohnSchwarzmantel,177–78.NewYork:ColumbiaUniversityPress.Xu,Yiqing.2017.“GeneralizedSyntheticControlMethod:CausalInferencewithInteractiveFixedEffectsModels.”PoliticalAnalysis25(1):57–76.Ziblatt,Daniel.2017.ConservativePoliticalPartiesandtheBirthofModernDemocracyinEurope.NewYork:CambridgeUniversityPress.JacobM.Grumbach984https://doi.org/10.1017/S0003055422000934 Published online by Cambridge University Press

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *