using electoral cycles in police hiring to estimate the effect of police on crime

《using electoral cycles in police hiring to estimate the effect of police on crime》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、江畔夏草UsingElectoralCyclesinPoliceHiringtoEstimatetheEffectofPoliceonCrimeBySTEVEND.LEVITT江畔夏草1.QuestionofInterest:thecausaleffectofpoliceoncrime.2.EndogenousProblemsSimultaneitybetweenpoliceandcrime:Evenifpolicereducecrime,a)Citieswithhighcrimeratesmaytendtohavelargepoliceforces.b)Citieswillhir

2、emorepolicewhencrimeisincreasing.Theendogenousproblemcausesapositivebias,whichevenshowsapositiverelationshipbetweenthetwoinpreviousliterature.3.EmpiricalEstimationMethods:IVInstrumentVariable:thetimingofmayoralandgubernatorialelections4.AssumptionsfortheEmpiricalEstimationMethodsInstrumentexog

3、eneity:electionsmustbeuncorrelatedwithcrime,exceptthroughvariablesthatareincludedintheequationexplainingcrime.Threats:electionsmightsystematicallyaffectthecrimeratethroughelectoralcyclesinothertypesofsocialspendingpoliticallyinducedfluctuationsineconomicperformanceSolution:controlforthesefac

4、tors:spendingoneducationandpublicwelfareprograms,stateunemploymentrate.5.DataUsedApanelof59U.S.citieswithobservationsrunningfrom1970-1992,satisfying:i.Thecitypopulationexceeds250,000atsomepointinthetimeperiodanalyzed;ii.Themayorisdirectlyelected.DataoncrimeandswornofficersaretakenfromtheFBIU

5、niformCrimeReports.7typesofcrime,categorizedintoviolentcrimeandpropertycrime.6.MainResults1)Policesubstantiallyreduceviolentcrime,buthaveasmallereffectonpropertycrime.2)2SLSestimatesareconsistentlymorenegativethanOLSestimates.7.ConclusionsAlthoughtheinstrumentvariablesestimatesareimprecise,si

6、nceelectioncyclesexplainonlyasmallfractionoftheoverallvariationinpolice,theresultsdoprovideevidencesuggestingthatadditionalpolicereducecrime.8.CommentsandThoughts1)Drawbacksa)PotentiallyimposeunwarrantedstructureThecoefficientsonthepolice,demographic,andstateandlocalspendingvariablesare江畔夏草con

7、strainedtobeidenticalforallviolentcrimeandallpropertycrime,whichmayimposeunwarrantedstructureonthedata.Theparameterrestrictionofdifferentcrimesmaynotbereasonable.2SLSresultsforviolentcrimemaybedrivenbytheestimateoftheeffectofpoliceonmur