Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
OrganizedCrimeandElectoralOutcomes.EvidencefromSicilyattheTurnoftheXXICentury
PaoloBuonanno GiovanniPrarolo PaoloVaninUniversityofBergamo UniversityofBologna UniversityofBologna
EuropeanJournalofPoliDcalEconomy,41,2017,61-74.
MoDvaDon
• RelaDonshipmafia-poliDcs:– Crucial,butunderinvesDgated– Whydoesorganizedcrimepersistsindemocracies,despiteitssociallyundersirableeffects?
• Mafiamay– CondiDonelectoraloutcomes(thispaper)– CondiDonelectedpoliDcians(notinthispaper)
ContribuDon• Focus
– Voteshares(VS)atItalianparliamentaryelecDons,1994-2008– DisaggregatedinfoonmafiapresenceandVSinSicilianmunicipaliDes– IVbasedondeterminantsofmafia’searlydistribuDon
• CanwefindevidenceofasystemaDclinkbetweentheSicilianmafiaandVSatpoliDcalelecDons?– Yes,ForzaItalia(SilvioBerlusconi’sparty)obtainedsignificantlyhigher
votesharesinmafia-plaguedmunicipaliDes
• CanweinterpretthiscorrelaDonasevidenceofmafia’sintervenDoninthemarketforvotes?– IVsuggeststhatmafia’spresenceraisedForzaItalia’svoteshares– Furtherevidenceneededtosayhowexactlythishappened
Literature• Italianmafia:
– NegaDveeffects:GDPloss(Pinob2015);lowerqualityofpoliDcians(GeysandDaniele2015);diversionofpublicfunds(BaroneandNarciso2015)
– Originsandpersistence:Gambefa1993,Bandiera2003,Pazzona2010,Buonannoetal2015)
• Marketforvotes:– BuyingelectedpoliDcians’vote:throughmarket(Pikefy1999,2000;
Dekeletal.2008);bribes(Snyder1991);lobbying(GrossmanandHelpman,1996;EstebanandRay2006)
– ControllingciDzens’vote:byemployers(BalandandRobinson2008);byarmedgroups(Acemogluetal.2013)
• MafiaandelectoraloutcomesinSicily:– 1946-1992:DeFeoandDeLuca(2013)– 1994-2008:ourpaper
Background• Italyintheearly1990s:newactorsbothinmafiaandin
poliDcs– CorrupDonscandalsandchangesininternaDonalaffairs– Maxi-trialagainstmafiainthelate1980sledbyjudgesFalconeand
Borsellino;furtherprosecuDonakertheywerekilledbythemafia
• MainnewpoliDcalactor:SilvioBerlusconi– FoundedForzaItaliain1993,primeministerin1994– PaidmafiaforprotecDoninthe1970s– HiredaSicilianmafioso,ViforioMangano,totendhishorses– MarcelloDell’Utri,Sicilianco-founderofForzaItalia,wassentencedto
7yearsformafiaconnecDonsin2014– Won61outof61parliamentaryseatsinSicilyin2001(VSinSicily
between30%and36%in1994-2006,47%in2008)– AntoninoGiuffrè,mafiabossturnedinformer:mafiasupportedForza
Italiasince1993,inexchangeforhelpinresolvingitsjudicialproblems
Data:MafiaandElectoralOutcomes• Contemporarymafiapresence:
– Seizedfirms:dummy=1ifatleastonefirmwasseizedtothemafiabefore2011(Law646,1982,“Rognoni-LaTorre”)
– Seizedrealestates:similardummyforland,apartments,etc– Seizures:maxoftheprevioustwodummies
• Laggedmafiapresence:– Mafia_1987:fromDeFeoandDeLuca(2013),dummybasedonamilitary
policereportin1987– Mafia_Cutrera:fromCutrera(1900),dummyfor337municipaliDes
• Electoraloutcomes:– ForzaItalia:averagevoteshare1994-2008(PopolodelleLibertàin2008)– Center-lek:averagevoteshare1994-2008
SicilianmunicipaliDeswithmafiaseizuresby2011
ForzaItalia’saveragevoteshareinSicilianmunicipaliDes,1994-2008
MafiainSicilianmunicipaliDes,fromCutrera(1900)
Data:Controls• Provincedummies
• Municipality-levelcontrolsfrom2001census– PopulaDondensity– ImmigraDonrate– ShareofworkingagepopulaDonwithatleasthighschool– Unemploymentrate
• GDPpercapitaatSLLlevelfrom2001census
• ShareofTVtaxevasion
• AveragealDtude(fordifferentinterestsandinstrumentvalidity)
Data:Instruments• Endogeneity:
– ReversecausaDon,omifedvariables,measurementerror
• Instruments:determinantsofearlymafiadiffusioninthelateXIXcentury,basedonBuonannoetal.(2015)– Numberofsulfurminesineachmunicipalityin1886,fromSquarzina(1963)• BoominXIXcentury,thendisappeared:temporaryincreaseindemandforprotecDonandopportuniDesforextorDon
– Agro-ecologicalsuitabilityforcereals,fromFAO-GAEZ• Importanceoflargehistoricallandholdings,reducedakerlandreformin1950(todaycerealsare8%ofculDvatedland,smallfracDonofGDP)
– DifferenceinalDtudewithinamunicipality,fromISTAT• LimitaDonsonagricultureandcaflebreeding,highercostofpolicing,hidingopportuniDesforcriminals
EmpiricalStrategy
• OLSesDmatesof:VSi=βmafiai+γʹXi+φp+εi
• Robustness– Differentsetsofcontrols– Differentmafiameasures,bothcontemporaryandhistorical– Inclusionof35electoraldistrictdummies(dropbigciDes)– Samplesplit1994-2001and2006-2008(electoralruleandpartymerge)– FalsificaDontest:center-lekvs.ForzaItalia– SpaDalesDmates:SEM,SAR,both
• Causality– IVesDmatesforbothForzaItaliaandcenter-lek
DescripDveStaDsDcs
PairwiseCorrelaDons
Table 2: Correlation matrix
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)Seized real estates (1) 1.0000Seized firms (2) 0.5750* 1.0000Seizure (3) 0.6229* 0.9733* 1.0000mafia 1987 (4) 0.3240* 0.2893* 0.2866* 1.0000mafia cutrera (5) 0.2290* 0.2041* 0.2097* 0.3257* 1.0000Pop. density (6) 0.2136* 0.2337* 0.2372* -0.0383 0.0519 1.0000Immigration (7) -0.0494 0.0610 0.0523 -0.0801 -0.2480* 0.0455 1.0000High school (8) 0.1566* 0.1933* 0.2008* -0.0859* -0.1114* 0.3922* 0.1589* 1.0000TV fee evasion (9) 0.0705 0.0763 0.0689 -0.0854* 0.0188 0.2233* 0.1060* -0.0032 1.0000Unemp. rate (10) 0.1337* 0.1936* 0.1859* 0.2734* 0.3774* 0.0587 -0.2295* -0.1200* 0.0531 1.0000GDP per capita (11) 0.2289* 0.1913* 0.1904* -0.0629 -0.0892 0.2972* 0.2209* 0.3451* 0.0621 -0.1059* 1.0000Caves (12) 0.1311* 0.1415* 0.1373* 0.1645* 0.1252* -0.0695 -0.0451 0.0838 -0.1177* 0.2555* 0.0983* 1.0000Difference in elevation (13) -0.0737 -0.1020* -0.0933* -0.0483 0.0828 -0.2633* -0.1660* -0.1231* 0.1091* -0.1331* -0.1738* -0.0845* 1.0000Cereals suitability (14) 0.2402* 0.2877* 0.2806* 0.2499* 0.2420* 0.0841 0.2331* 0.0499 -0.1464* 0.1447* 0.0300 0.0702 -0.4002* 1.0000
Notes. This table presents pairwise correlations between all variables. * denotes significance at 1%.
20
ForzaItaliawithContemporaryMafia
ForzaItaliawithHistoricalMafia
Center-lekwithContemporaryMafia
Center-lekwithHistoricalMafiaTable 6: The effect of historical mafia on the share of votes of the centre-left coalition
(1) (2) (3) (4) (5) (6)
mafia 1987 0.0106 -0.0053 -0.0030[0.011] [0.010] [0.009]
mafia cutrera 0.0178 0.0041 0.0083[0.011] [0.031] [0.019]
Pop. density -0.0118** -0.0197*[0.005] [0.010]
Immigration rate -0.0028 -0.0002[0.005] [0.004]
High school -0.0572 -0.0916[0.163] [0.185]
TV fee evasion -16.0909*** -15.4205***[3.851] [3.154]
Unemp. rate 0.0701 0.0307[0.367] [0.334]
GDP per capita -0.0026* -0.0016[0.001] [0.001]
Average altitude 0.0158 0.0103[0.028] [0.031]
Prov. FE N Y Y N Y Y
Observations 375 375 374 337 337 336R-squared 0.002 0.181 0.254 0.008 0.202 0.269
Notes. This table presents the results of OLS estimates. The dependent variable is the average share of votesobtained at national elections by the centre-left coalition over the period 1994-2008. The two main explanatoryvariables are dummies indicating whether mafia was active according to a 1987 military police report to a par-liamentary committee (Comando Generale dell’Arma dei Carabinieri, 1987) and according to Cutrera (1900),respectively. The other control variables are described in the main text. Robust standard errors (clustered atthe province level when province dummies are introduced) are presented in parentheses. *, ** and *** de-note rejection of the null hypothesis of the coefficient being equal to 0 at 10%, 5% and 1% significance level,respectively.
24
SpaDalEsDmates:ForzaItaliaTable 7: Spatial estimates: share of votes for Forza Italia
(1) (2) (3) (4) (5) (6)
Seizure (total) 0.0353*** 0.0267*** 0.0263*** 0.0365*** 0.0246*** 0.0247***[0.009] [0.008] [0.008] [0.008] [0.008] [0.008]
Pop. density 0.0214*** 0.0186** 0.0188** 0.0219*** 0.0186** 0.0194**[0.008] [0.008] [0.008] [0.008] [0.008] [0.008]
Immigration rate 0.0057* 0.0037 0.0037 0.0057* 0.0045 0.0045[0.003] [0.003] [0.003] [0.003] [0.003] [0.003]
High school -0.0159 -0.0199 -0.0240 -0.0183 -0.0218 -0.0269[0.099] [0.095] [0.095] [0.099] [0.094] [0.093]
TV fee evasion 4.0085 5.2475 5.1400 4.4419 4.9333 5.6057[3.918] [3.886] [3.885] [3.954] [3.930] [3.936]
Unemp. rate 0.0821 -0.0552 -0.0661 0.0937 -0.0168 -0.0129[0.198] [0.215] [0.214] [0.199] [0.222] [0.221]
GDP per capita 0.0021* 0.0024** 0.0023* 0.0022* 0.0026** 0.0026**[0.001] [0.001] [0.001] [0.001] [0.001] [0.001]
Average altitude -0.0258* -0.0450*** -0.0461*** -0.0228* -0.0307** -0.0310**[0.014] [0.017] [0.017] [0.013] [0.015] [0.015]
� 0.0027 0.0027 -0.0191 -0.0501[0.004] [0.005] [0.038] [0.040]
⇢ 0.0857*** 0.0848*** 0.4328*** 0.4694***[0.011] [0.011] [0.061] [0.066]
Observations 382 382 382 382 382 382Notes. This table presents the results of a spatial model estimated by means of the generalised spatial two stage leastsquares (GS2SLS) estimator of Kelejian and Prucha (1998), with the most complete specification used in Table 3 andusing total seizures to measure contemporary mafia presence. Columns 1 to 3 employ a non-standardised contiguitymatrix, while a row-standardised one is used in columns 4 to 6. A Spatial Error model, a Spatial Autoregressive modeland a model that combines the two by considering both a spatial lag and a spatial error structure are respectively presentedin columns 1 and 4, columns 2 and 5 and columns 3 and 6. � is the spatial error term, while ⇢ is the spatial lag. Thedependent variable is the average share of votes obtained at national elections by Forza Italia over the period 1994-2008. Robust standard errors (clustered at the province level when province dummies are introduced) are presented inparentheses. *, ** and *** denote rejection of the null hypothesis of the coefficient being equal to 0 at 10%, 5% and 1%significance level, respectively.
25
IVEsDmates:ForzaItaliaTable 8: IV estimates: mafia on share of votes of Forza Italia
(1) (2) (3) (4) (5)
Seized real estates 0.1314**[0.055]
Seized firms 0.1017**[0.050]
Seizure (total) 0.1298**[0.054]
mafia 1987 0.1735**[0.080]
mafia cutrera 0.1496***[0.057]
Pop. density 0.0167* 0.0184** 0.0149 0.0250** 0.0289***[0.009] [0.008] [0.009] [0.010] [0.010]
Immigration rate 0.0031 0.0073** 0.0035 0.0040 0.0066[0.004] [0.003] [0.004] [0.003] [0.004]
High school -0.2314 -0.0657 -0.2308 0.0522 0.1018[0.161] [0.118] [0.159] [0.116] [0.124]
TV fee evasion 1.7910 2.4865 1.8591 5.5610 3.8373[5.776] [5.469] [5.713] [5.857] [5.608]
Unemp. rate -0.0660 0.1795 -0.0232 -0.1701 0.0246[0.245] [0.247] [0.236] [0.329] [0.267]
GDP per capita 0.0007 0.0002 0.0007 0.0020 0.0006[0.002] [0.002] [0.002] [0.001] [0.001]
Average altitude -0.0160 -0.0191 -0.0192 -0.0283 -0.0310*[0.017] [0.016] [0.017] [0.018] [0.019]
Observations 381 381 381 373 335
Notes. This table presents the results of the second stage of the IV estimates. The dependent variable is theaverage share of votes obtained at national elections by Forza Italia over the period 1994-2008. The instru-mented explanatory variable in column 1(2 and 3, respectively) is a dummy indicating whether by 2011 atleast one firm (real estate, or at least one of either, respectively) were seized to the mafia in a given munici-pality (source: Agenzia del Demanio). In column 4 and 5 the instrumented explanatory variables are dummiesindicating whether mafia was active according to a 1987 military police report to a parliamentary committee(Comando Generale dell’Arma dei Carabinieri, 1987) and according to Cutrera (1900), respectively. The othercontrol variables are described in the main text. Province fixed effects are always included. Robust standarderrors clustered at the province level are presented in parentheses. *, ** and *** denote rejection of the nullhypothesis of the coefficient being equal to 0 at 10%, 5% and 1% significance level, respectively.
26
IVEsDmates:FirstStage
IVEsDmates:Center-lek
SummaryandConclusion• Mainresult:ForzaItaliaobtained4%-5%highervoteshares
between1992and2008inSicilianmunicipaliDesplaguedbymafia– Robusttomafiameasures,controls,spaDalesDmates
• IVsuggeststhat,coherentlywithtesDmoniesbyformermafialords,thisislikelytoreflectanexplicitchoicebymafiafamiliestobackForzaItaliainexchangeformorefavorablepolicies– Noneoftheseresultsholdforcenter-lek
• InstrumentedmafiaraisesForzaItalia’svoteshareby13%– DidBerlusconipayoff(forthemafia)?
• ContribuDon:– Evidencethathistoricallypredictedmafiapresenceraisesaspecific
party’svotesharesatpoliDcalelecDons– MayhelpexplainwhyelectedpoliDciansarereluctanttofightmafia
anddestroyasourceofvotes