Are traffic violators criminals? searching for answers in experiences of European countries

1 ARE TRAFFIC VIOLATORS CRIMINALS? SEARCHING FOR ANSWERS IN EXPERIENCES OF EUROPEAN COUNTRIES 1. INTRODUCTION. Road a ic acciden s (RTAs) a e a majo cause o mo ali y and se ious mo bidi y wi h high cos s along he en i e Eu opean Union (EU) geog aphy (Cas illo-Manzano e al., 2014; O si e al., 2012). Wo ldwide, among he issues ha con ibu e o he occu ence o RTAs, mos o s udies highligh beha io al ac o s as a p incipal explana ion; pa icula ly hose ha educe capabili ies (d owsiness, a igue, alcohol o d ug consump ion) o inc ease isk while d i ing ( a ic o ences, d i e s ess, iolence and agg essi eness in a ic, mo o ehicle c imes) (Pe idou & Mous aki, 2000 analyze his classi ica ion in dep h). A e ini ial s udies such as Po e ield (1960), in he las wo decades special a en ion has been paid in he li e a u e o he connec ion be ween oad a ic sa e y and c iminal beha io . The oles o bo h a ic law en o cemen (TLE) and legal sys ems ha e been highligh ed as de e ence and puni i e measu es o imp o e compliance wi h a ic ules (see he comp ehensi e in e na ional o e iews p o ided by B ace e al., 2009 and Zaal, 1994, espec i ely). Much esea ch has p o ided e idence wo ldwide ha people wi h an isocial a i udes who commi agg essi e and o he o ences due o a lack o assimila ion o social no ms, a e mo e equen ly in ol ed in isky si ua ions in gene al (see e.g. Fische and Poland, 1998 on how social exclusion may e en se e as a con ol mechanism when adi ional means o social con ol such as punishmen s o de e ence a e ine ec i e o insu icien ) and a e speci ically mo e p one o RTAs and a ic iola ions (B ough on, 2007; Săucan e al., 2012; Süme , 2003; Wes e al., 1993). 2 In ac , he ela ionship be ween oad sa e y and c iminali y has been add essed by applying di e en psychological, sociological and c iminological esea ch app oaches (B ace e al., 2009; Junge e al., 1995a, o a b oad analysis). Fo ins ance: he endency o commi c ime acco ding o pe sonali y ai s (wi h he heo y labeled as “Mild Social De iance” based on in ol emen in acciden s associa ed wi h exp ession o hos ile a i udes, agg ession and an isocial o ms o beha io , Meadows e al., 1998; Wes e al., 1993); di e se si ua ional hypo heses ha sugges ha he way one d i es is closely linked o o he beha io al ai s in one’s pe sonal and social li e, because d i ing beha io is ac ually a pa o a complex sys em de e mined by a i udinal and mo i a ional ac o s ha a ec indi iduals in hei ela ions wi h socie y ( he so-called “Sel -con ol Theo y” and “Social Con ol Theo y” and he “Hie a chical App oach” applied o a ic esea ch by Ha akka e al., 2002; Junge e al., 1995b; Junge & T emblay, 1999, among o he s). Howe e , he connec ion be ween c iminal o unsa e beha io and oad sa e y seems o be a ela i ely unde - esea ched opic bo h na ionally and abo e all in e na ionally. Di e en causal links a e explo ed in he li e a u e, p edominan ly: be ween gene al c imes and a ic o ences wi h an special emphasis in ecidi ism (B ough on, 2007; Palk e al., 2005; Rose, 2000); be ween an isocial o nega i e beha io (homicide, obbe y, consump ion o alcohol and illici d ugs) and dange ous d i ing (E e e e al., 1999; Ma shall e al., 1996); and o a mino deg ee, he ela ionship be ween gene al c iminal beha io and RTAs (in e ms o he numbe o acciden s, a ali ies o inju y a es) (Giacopassi & Fo de, 2000; Po e ield, 1960). Al hough he e is a gene al consensus on he main heo e ical hypo heses ha a s ong co ela ion holds, ce ain au ho s s a e ha his migh be spu ious and ha he s a is ical associa ion be ween c ime and RTAs is ac ually he esul o a common causal p ocess ela ed 3 o socioeconomic con ex (age and sex, isk exposu e o acciden , income le el, unemploymen , e c.), (Junge e al., 1995a, 1995b; Junge & T emblay, 1999; Si ak, 1983). Ou con ibu ion o he li e a u e alls in o his ield. The main aim o he cu en s udy is o examine econome ically he associa ion be ween c iminal beha io and a ic a ali ies pe capi a h ough a panel o he 28 cu en membe s a es o he Eu opean Union (EU28) o he 1999-2010 pe iod. We aim o es he hypo hesis ha c ime a es (and speci ically, mo o ehicle- ela ed c imes) may be conside ed as p edic o s o a al RTAs. Following B ace e al. (2009) and Giacopassi & Fo de (2000), he e has been e y li le esea ch unde aken o explo e he e ec s o he ela ionship be ween gene al c iminal beha io and c ash in ol emen speci ically. We conside ha ou esea ch has se e al ad an ages o e p e ious s udies ha gua an ee i s o iginali y. Fi s ly, his issue has ecei ed some a en ion om esea che s in ce ain coun ies ha ha e analyzed he issue a he na ional le el; al hough, o ou knowledge, no p e ious s udy conside s a b oad panel o coun ies, o ocuses speci ically on he in luence o c iminal beha io (no only gene al c ime bu also speci ic mo o ehicle c ime) on dea h om a ic acciden s. Secondly, ou s udy also aims o explo e he ole o he legal sys em on he ela ionship be ween c iminali y a es and RTAs om a geog aphical poin o iew, o he EU28 coun ies. Logically, he EU28 coun ies show a he e ogeneous spec um o cul u al a i udes and social no ms (Ve eeck & V olix, 2007), and also a wide ange o a ic law amewo ks (e.g., se e al coun ies wi h be e oad sa e y eco ds, such as he Ne he lands, p ocesses a ic o enses unde adminis a i e laws, while o he leade s such as he Uni ed Kingdom apply a c iminal a ic law sys em; ETSC, 1999). In ac , some s udies s a e ha compliance 4 wi h he a ic ules (and wi h gene al law) in each coun y, is dependen no only on he social a i udes o oad use s o ce ain beha io , bu also on he op imiza ion o he law and i s en o cemen (TLE o a ic) ha should o m pa o a global s a egy o imp o e oad sa e y: legisla ion, educa ion, en o cemen and judicial p ocedu es (see e.g. Wegman, 1992). The e ec s o changes in he le el o sanc ions ( icke s and ines, jail sen ences, deme i /penal y poin s sys ems) o en o cemen (numbe o police o ice s, numbe o b ea h es s o alcohol/d ugs, speed came as) on he le el o oad sa e y (acciden a es) has been analyzed by much economic esea ch in e es ed in an op imal alloca ion o sca ce esou ces, bu wi h no conclusi e indings. Ne e heless, he e seems o be gene al ag eemen ha en o cemen o oad ules should be aimed p ima ily a de e ence, being accompanied by publici y ha is unp edic able and di icul o a oid, wi h a mix u e o highly isible and less isible ac i i ies and con inued o e a long pe iod o ime because he e ec s a e sho e m (S anoje ić e al., 2012; Zaal, 1994). Special a en ion has been paid by au ho s such as Deshap iya & Iwase (1996) and Tay (2005, 2010) o he se e i y o sanc ions as a de e ence mechanism o inc ease he indi idual’s pe cei ed isk o app ehension and punishmen . In he oad sa e y a ena, s ic e sanc ions seem o achie e be e and as e esul s i hey a e mainly de e mined by he legisla u e and adminis a i e/judicia y sys em (i.e., he in oduc ion o oughe jail sen ences and punishmen s may educe a ic iola ion a es, al hough o a limi ed pe iod; see o example cases as a ied as Cas illo-Manzano e al., 2011 o Spain; Li e al., 2014 o Hong Kong; Libe a i e al., 2001 o B azil; Mon ag, 2014 o he Czech Republic o Sen, 2001 o Canada). The e o e, in his pape we examine he na u e o he mos se e e legal sys ems as de e mining ac o s o he le el o EU28 oad sa e y. In sho , we a emp o de e mine i he e may be a clea co ela ion be ween oad a ali y a es and he mains eam ypes o c ime a an in e na ional le el, which would se e o jus i y, a leas p ima acie, he inc easingly widesp ead endency o conside a ic o ences as 5 c imes and oughen he penal ies imposed in he penal codes. Following his a gumen , we simul aneously analyze whe he he deg ee o se e i y o he legal sys em applied o a ic o ences could be a good ool in he cons an s uggle o Wes e n socie ies o diminish he d ama o RTAs. A e his heo e ical in oduc ion, he ollowing sec ions p esen he me hodological amewo k o ou esea ch, a discussion o ou indings and inally a summa y o ou main conclusions. 2. VARIABLES AND METHODOLOGY. In ou empi ical analysis, we de elop a ixed-e ec s model ha akes he ollowing o m o coun y i du ing pe iod : Yi = α + βkXi + γkZi + λkWi + μi + ν + εi (1) whe e Yi is he log o he o al pe capi a a ali y a e (wi hin 30 days o he acciden , acco ding o he Vienna Con en ion de ini ion) 1 , Xi e e s o a iables ha iden i y he c iminal a i udes o he coun y’s inhabi an s, Zi con ains he ec o o he coun y’s economic and demog aphic a ibu es, and Wi a e bo h he se e i y o legal sys em- ela ed a iables as p oxies o law en o cemen , and speci ic a iables ela ed o oad sa e y policies. μi a e coun y ixed e ec s ha con ol o omi ed ime-in a ian coun y-speci ic a iables, ν a e yea dummies ha con ol o he common end in all he coun ies in he da ase and εi is a mean-ze o andom e o . 1 Albala e (2008), Albala e & Bel (2011), Dee (2001) and Eisenbe g (2003) conside ha his is he mos app op ia e dependen a iable o assessing oad a ic a ali ies as he in e p e a ion o policy a iables is clea e . 6 We use da a o he EU28 coun ies o he ime pe iod om 1999 o 2010. This means ha his is he b oades da abase in e ms o bo h ime and geog aphy used in an analysis wi h hese objec i es. Table 1 p o ides he desc ip ions, in o ma ion sou ces and desc ip i e s a is ics (mean and s anda d de ia ion) o all he a iables used in he analysis. [INSERT TABLE 1] As we s a e abo e, he main aim o ou empi ical analysis is o examine whe he he c iminal a i udes o he inhabi an s o a coun y ha e an e ec on a ic a ali y a es. Acco ding o ea lie s udies (B ace e al., 2009; Junge e al., 1995a; Junge e al., 2001b; Po e ield, 1960; Si ak, 1983; Süme , 2003), we may expec a posi i e co ela ion be ween c iminal ac i i ies and he a i udes o d i e s on he oad, in he sense o a highe likelihood ha hey a e in ol ed in a a al RTA. The hypo hesis we es he e is ha le els o social agg ession a e a p edic o o a ic acciden s: highe c ime a es a e associa ed wi h mo e dea hs on he oads, bea ing in mind he possible in luence o he spu iousness hesis suppo ed by schola s such as Junge e al. (1995a) and Si ak (1983), as we men ioned in he In oduc ion abo e. In equa ion (1) we include ou explana o y a iables ha iden i y he c iminali y a es o a coun y’s popula ion, and we also conside some con ol a iables ha a e desc ibed in he ollowing pa ag aphs. We collec ed comple e da a o wo a iables ela ed wi h gene al c iminali y a es, also used by p e ious empi ical esea ch, al hough i only conside ed speci ic coun ies (Giacopassi & Fo de, 2000; Junge e al., 1995b; Junge e al., 2001a; Si ak, 1983): he numbe o homicides and he numbe o obbe ies. Mo eo e , as we men ion abo e in he In oduc ion, almos all 7 he ea lie esea ch ocused on gene al c ime a iables o on ypical d i ing iola ions- ela ed a iables, such as alcohol consump ion (Deshap iya & Iwase, 1996; Pe ie e al., 2010; Wes e al., 1993) o o he s, such as so-called “joy- iding” c ime (Ma shall e al., 1996). By o he hand, au ho s such as B omley and Nelson, 2002, ind ha alcohol is no ed as a con ibu o y ac o in eco ded c imes. Fu he mo e, we include wo a iables o c imes speci ically linked o oad ehicle a ic: he numbe o d ug a icking c imes and he numbe o mo o ehicle he s. All hese a iables a e measu ed in ela i e e ms acco ding o he coun y’s popula ion. While we may expec ha he coe icien associa ed wi h he c ime a iables is posi i e, such an e ec should be pa icula ly ele an o hose a iables, such as d ugs, ha a e mo e linked o oad ehicle a ic, in ag eemen wi h Gje de e al. (2011) and Smink e al. (2005) o ce ain coun ies. Along wi h a iables ela ed wi h di e en ypes o c ime in he coun y’s popula ion, we also analyze whe he he se e i y o he c iminal jus ice sys em in luences oad mo ali y a ios. As a p oxy o he se e i y o he penal/legal sys em in a coun y, we include an index a iable ha akes he alue ze o i no li e imp isonmen sen ence may be applied, he alue one i such a sen ence is e isable and he alue 2 i he inde ini e sen ence is no e isable. Ou expec a ion is ha he coe icien associa ed wi h his a iable will be nega i e, ollowing p io esul s o e.g., Cas illo-Manzano e al. (2011), Libe a i e al. (2001) Mon ag, 2014 o Sen (2001) ega ding he bene icial e ec s o s ic e punishmen applied o a ic iola o s, speci ically on oad sa e y ou comes. Indeed, he a ic a ali y a es may be lowe in coun ies wi h mo e se e e penal/legal/adminis a i e sys ems as he penal sys ems may ha e a s ong in luence on he social beha io o ci izens and d i e s. 8 We also conside addi ional a iables o oad sa e y policies. Following p e ious s udies on oad sa e y (Albala e, 2008; Cas illo-Manzano & Cas o-Nuño, 2012; Cas illo-Manzano e al., 2013, 2014; Deshap iya & Iwase, 1996; Eisenbe g, 2003; McCa hy, 2005), we conside he in luence o p e en i e policies such as legal blood alcohol limi s, and include a dummy a iable ha akes a alue o one o coun ies and pe iods whe e he maximum blood alcohol concen a ion allowed is lowe han 0.5 g/L. Mos EU28 coun ies ha e se he limi a 0.5 o lowe , so we a e able o es whe he blood alcohol concen a ion a es lowe han 0.5 a e e ec i e in educing oad a ic a ali ies. Addi ionally, we include a a iable ha cap u es he applica ion o poin s-based d i ing licenses. Indeed, one o he explana o y a iables in equa ion (1) is an index a iable ha akes he alue one i a penal y d i ing license sys em is applied, he alue wo i a deme i d i ing license sys em is applied and he alue ze o i any poin s sys em is applied. We expec o ind a nega i e sign in he coe icien associa ed wi h his a iable. The in oduc ion and applica ion o any poin s sys em o d i ing licenses may lead o lowe a ic a ali y a es (i.e. he s ong de e ence e ec analyzed by De Paola e al., 2013 o I aly), al hough he long- e m e ec o his policy has been ques ioned (see Cas illo-Manzano & Cas o-Nuño, 2012, o an in e na ional me a-analysis o he e ec s o such policies). Ano he oad a ic policy a iable ha we ake in o accoun is o he maximum speed limi allowed on mo o ways. Acco ding o p e ious s udies (A ukaa , 2003; El ik, 2012), we may expec a posi i e sign in he coe icien associa ed wi h his a iable, as a highe speed limi (and i s subsequen en o cemen ) may ha e as a consequence highe oad a ic a ali ies. 9 Wi h espec o con ol ac o s, we include a iables ha a e ypically used in s udies o oad a ic sa e y and ha a e ela ed o he coun y’s economic condi ions. In his ein, we include he pe capi a g oss domes ic p oduc (GDP) as an indica o o he coun y’s economic de elopmen (Cas illo-Manzano e al., 2013, 2014; Ga cía-Fe e e al., 2007; G imm & T eibich, 2012; O si e al., 2012). The ela ionship be ween a coun y’s economic de elopmen and oad a ic a ali ies is complex and we may expec ha i akes a non-linea o m (Anba ci e al., 2006; Bishai e al., 2006; Kopi s & C oppe , 2005). Indeed, a ali y a es may inc ease wi h economic de elopmen in e y poo coun ies due o inc eased exposu e o RTAs. Howe e , he ela ionship be ween economic de elopmen and a ic a ali y a es may become la o e en in e a e eaching a ce ain weal h h eshold. We es he hypo hesis conce ning non- linea i y be ween GDP and a ic a ali ies by including he GDP and he squa e o he GDP as explana o y a iables. I he non-linea ela ionship is con i med, hen he coe icien associa ed wi h he GDP a iable will be posi i e and he coe icien associa ed o he squa e o he GDP a iable will be nega i e. Meanwhile, s udies show ha o e all c ime a es and p ope y c imes a e highe when economic condi ions a e poo due o unemploymen o ecession (K üge , 2011) wi h a posi i e ela ion e idenced o he EU by Al indag (2012). O he s udies e en sugges how income inequali y may play an impo an ole in he de e mina ion o he c ime a e (Sachsida e al., 2010), ollowing he a ional c ime heo y, because he oppo uni y cos o commi ing a c ime seems o dec ease and he ne indi idual’s bene i migh be highe (K üge , 2013). GDP is he e o e qui e a use ul con ol a iable when he in en ion is o s udy any possible co ela ion wi h c ime a es. 16 s udy is o a b oad sample: EU28 and a ull wel e yea s o esea ch, and shows ha his is no a gene ic ela ionship bu is limi ed o speci ic c iminal beha io being linked o eckless d i ing. These a e speci ically he o ense ca ego ies o mo o ehicle he and d ug a icking, which closely co ela e wi h a ic acciden s. Ano he ele an esul o ou esea ch is ha he e is also a clea co ela ion be ween a educ ion in he oad a ali y a e and he li e imp isonmen a iable, which can be conside ed o be a clea p oxy o he gene al s ingency o each coun y’s c iminal code. Thus, he se e i y o he sanc ion may be used as a de e ence mechanism o inc ease he indi idual’s pe cei ed isk o app ehension and punishmen (a signal abou he isibili y o s ic e TLE and hea ie sanc ions); and social con ol migh be achie ed h ough de e ence by s ic e laws. All hese commen s suppo he idea o a clea co ela ion be ween c iminal beha io and RTAs, inasmuch as speci ic s a egies used o con end wi h he o me also help o educe RTAs. The oad a ali y a es seem o be highe in Eu opean coun ies whe e he inhabi an s show mo e socially agg essi e beha io , while hey a e lowe in coun ies wi h mo e se e e penal sys ems. Finally, oge he all hese indings could be conside ed o be indi ec empi ical suppo o coun ies such as B azil, Canada, he Czech Republic o Spain ha ha e oughened sanc ions o comba oad acciden s, ei he h ough hea ie economic and adminis a i e sanc ions o e en jail sen ences, which a e en isaged in hei penal codes o eckless beha io behind he wheel. In sho , i shows suppo o he coun ies ha ea his eckless beha io and he people esponsible o i as c iminals. 17 REFERENCES A ukaa , F. K. (2003): ‘Speed con ol in de eloping coun ies: issues, challenges and oppo uni ies in educing oad a ic inju ies’, Inju y Con ol Sa e y P omo ion, 10, 77– 81. Albala e, D. (2008): ‘Lowe ing Blood Alcohol Con en Le els o Sa e Li es: The Eu opean Expe ience’, Jou nal o Policy Analysis and Managemen , 27, 20–39. Albala e, D. and G. Bel (2012): ‘Mo o ways, olls, and oad sa e y: E idence om Eu ope’, SERIES-Jou nal o Spanish Economic Associa ion, 3, 457-473. Al indag, D. T. (2012): ‘C ime and unemploymen : E idence om Eu ope’, In e na ional Re iew o Law and Economics, 32,145– 157. Anba ci, N., Escale as, M. and C. Regis e (2006): ‘T a ic Fa ali ies and Public Sec o Co up ion’, Kyklos, 59, 327-344. Anwaa , A., Anas asopoulos, P., Ping Ong, G., Labi, S. and M. Bin Islam (2012): ‘Fac o s a ec ing highway sa e y, heal h ca e se ices, and mo o iza ion. An explo a o y empi ical analysis using agg ega e da a’, Jou nal o T anspo a ion Sa e y and Secu i y. 4, 94-115. B ace, Ch., Whelan, M., Cla k, B. and J. Oxley (2009): The ela ionship be ween c ime and oad sa e y. Monash Uni e si y Acciden Resea ch Cen e. Repo Documen No. 284. Aus alia. B omley, R. D., and Nelson, A. L. (2002): ‘Alcohol- ela ed c ime and diso de ac oss u ban space and ime: e idence om a B i ish ci y’. Geo o um, 33, 239-254. B ough on, J. (2007): ‘The co ela ion be ween mo o ing o ences and o he ypes o o ences’, Acciden Analysis and P e en ion, 39, 274-283. 18 Bishai, D., Qu esh, A., James, P. and A. Gha a (2006): ‘Na ional oad casual ies and economic de elopmen ’, Heal h Economics, 15, 65–81. B a e , E. R. and R.E. T empel (2004): ‘A e olde d i e s ac ually a highe isk o in ol emen in collisions esul ing in dea hs o non- a al inju ies among hei passenge s and o he oad use s?’ Inju y P e en ion, 10, 27-32. Cas illo-Manzano, J. I., Cas o-Nuño, M. and D.J. Ped egal (2011): ‘Can ea o going o jail educe he numbe o oad a ali ies? The Spanish expe ience’, Jou nal o Sa e y Resea ch, 42, 223-228. Cas illo-Manzano, J.I. and M. Cas o-Nuño (2012): ‘D i ing licenses based on poin s sys ems: E icien oad sa e y s a egy o la es ashion in global anspo policy? A wo ld wide me a-analysis’, T anspo Policy, 21, 191–201. Cas illo-Manzano, J. I., Cas o-Nuño, M. and X. Fageda (2014): ‘Could being in he Eu opean Union sa e li es? An econome ic analysis o he Common Road Sa e y Policy o he EU- 27’, Jou nal o Eu opean Public Policy, 21, 211-229. Cas illo-Manzano, J.I., Cas o-Nuño, M. and X. Fageda (2013): ‘Can heal h public expendi u e educe he agic consequences o oad a ic acciden s? The EU-27 expe ience’, Eu opean Jou nal o Heal h Economics, 15, 645-652. De Paola, M., Scoppa, V. and M. Falcone (2013): ‘The de e en e ec s o he penal y poin s sys em o d i ing o ences: a eg ession discon inui y app oach’, Empi ical Economics, 45, 965–985. Dee, T. S. (2001): ‘Does se ing limi s sa e li es? The case o 0.08 BAC laws’, Jou nal o Policy Analysis and Managemen , 20, 111-118. 19 Deshap iya, E.B.R. and N. Iwase (1996): ‘A e lowe legal blood alcohol limi s and a combina ion o sanc ions desi able in educing d unken d i e -in ol ed a ic a ali ies and a ic acciden s?’, Acciden Analysis and P e en ion, 28,721-731. Eisenbe g, D. (2003): ‘E alua ing he e ec i eness o policies ela ed o d unk d i ing’, Jou nal o Policy Analysis and Managemen , 22, 249-274. El ik, R. (2012): ‘Speed limi s, en o cemen and heal h consequences’, Annual Re iew o Public Heal h, 33, 225-238. ETSC (Eu opean T anspo Sa e y Council) (1999): Police en o cemen s a egies o educe a ic casual ies in Eu ope. B ussels. E e e , S.A., Low y, R., Cohen, L.R. and A.M. Dellinge (1999): ‘Unsa e mo o ehicle p ac ices among subs ance-using college s uden s’, Acciden Analysis and P e en ion, 31, 667–673. Fische , B., and Poland, B. (1998): ‘Exclusion,‘ isk’, and social con ol— e lec ions on communi y policing and public heal h’. Geo o um, 29, 187-197. Ga cía-Fe e , A., De Juan, A. and P. Poncela (2007): ‘The ela ionship be ween oad a ic acciden s and eal economic ac i i y in Spain: Common cycles and heal h issues’, Heal h Economics, 16, 603-626. Giacopassi, D. and D.R. Fo de (2000): ‘B oken windows, c umpled ende s, and c ime’, Jou nal o C iminal Jus ice, 28, 397-405. Gje de, H., No mann, P.T., Ch is ophe sen, A.S., Samuelsen, S.O. and J. Mø land (2011): ‘Alcohol, psychoac i e d ugs and a al oad a ic acciden s in No way: A case-con ol s udy’, Acciden Analysis and P e en ion, 43, 1197-1203. G imm, M. and C. T eibich C (2012): ‘De e minan s o oad a ic c ash a ali ies ac oss Indian S a es’, Heal h Economics, doi: 10.10002/hec.2870. 20 Ha akka, M., Keskinen, E., G ege sen, N. P., Glad, A. and K. He ne koski (2002): ‘F om con ol o he ehicle o pe sonal sel -con ol; b oadening he pe spec i es o d i e educa ion’, T anspo a ion Reseach Pa F-Psychology and Beha io , 5, 201-215. Jam oz, K. (2012): ‘The impac o oad ne wo k s uc u e and mobili y on he na ional a ic a ali y a e’, P ocedia-Social and Beha io al Sciences, 54, 1370-1377. Junge , M., Te louw G.J. and P.G.M. Van de Heijden (1995a): C ime and acciden in ol emen in young oad use s. In G.B. G ayson (Ed.). Beha iou al Resea ch in Road Sa e y V, 35-54. C ow ho ne: T anspo a ion and Road Resea ch Labo a o y. Junge , M., Te louw, G.J. and P.G.M. Van de Heijden (1995b): ‘C ime, acciden s and social con ol’, C iminal Beha io and Men al Heal h, 5, 386–410. Junge , M. and R.E. T emblay (1999): ‘Sel -con ol, acciden s, and c ime’, C iminal Jus ice and Beha io , 26, 485-501. Junge , M., Wes , R. and R. Timman (2001a): ‘C ime and isky beha io in a ic: an example o c oss-si ua ional consis ency’, Jou nal o Resea ch C ime and Delinquency, 38, 439-459. Junge , M., Van de Heijden, P. and C. Keane C (2001b): ‘In e ela ed ha ms: Examining he associa ions be ween ic imiza ion, acciden s, and c iminal beha io ’, In e na ional Jou nal o Inju y Con ol, 8, 13-28. Kılıç R. and P. McCa hy (2012): ‘Long- un equilib ium and sho - un dynamics be ween isk exposu e and highway sa e y’, Empi ical Economics, 42, 899–913. Kopi s, E. and M. C oppe (2005): ‘T a ic a ali ies and economic g ow h’, Acciden Analysis and P e en ion, 37, 169-178. K üge , N.A. (2011): ‘The Impac o Economic Fluc ua ions on C ime – A Mul iscale Analysis’, Applied Economics Le e s, 18, 179–182. 21 K üge , N.A. (2013): ‘Fa al connec ions−socioeconomic de e minan s o oad acciden isk and d unk d i ing in Sweden’, Jou nal o Sa e y Resea ch, 46, 59–66. Lang o d, J., Hakamies-Blomq is , L. and R. Me ho s (2006): ‘Olde d i e s do no ha e high c ash isk – a eplica ion o low mileage bias’, Acciden Analysis and P e en ion, 38, 574–578. Le in, A., Lin, C. F. and J.C. Chia-Shang (2002): ‘Uni Roo Tes s in Panel Da a: Asymp o ic and Fini e Sample P ope ies’, Jou nal o Econome ics, 108, 1-24. Li, H., G aham, D. J. and A. Majumda (2012): ‘The e ec s o conges ion cha ging on oad a ic casual ies: A causal analysis using di e ence-in-di e ence es ima ion’, Acciden Analysis and P e en ion, 49, 366-377. Li, Y.C., Sze, N.N. and S.C. Wong (2014): ‘E ec i e measu es o comba d ink d i ing o ences: an a i udinal model in Hong Kong’, T anspo me ica, doi:10.1080/23249935.2013.854425. Libe a i, C.L.B., And ade, S. M. and D.A. Soa es (2001): ‘The new B azilian a ic code and some cha ac e is ics o ic ims in sou he n B azil’, Inju y P e en ion, 7, 190-193. Ma shall, C., Boyd, K.T. and C.G. Mo an (1996): ‘Inju ies ela ed o ca c ime: he joy- iding epidemic’, Inju y, 27, 79-80. McCa hy, P. (2005): ‘Alcohol, public policy and highway c ashes’, Jou nal o T anspo Economics and Policy, 39, 109-125. Meadows, M. L., S adling, S. G. and S. Lawson (1998): ‘The ole o social de iance and iola ions in p edic ing oad a ic acciden s in a sample o young o ende s’, B i ish Jou nal o Psychology, 89, 417-431. Mon ag, J. (2014): ‘A adical change in a ic law: e ec s on a ali ies in he Czech Republic’, Jou nal o Public Heal h, doi: 10.1093/pubmed/ du005. 22 Noland, R. B. (2003): ‘T a ic a ali ies and inju ies: The e ec o changes in in as uc u e and o he ends’, Acciden Analysis and P e en ion, 35, 599-611. O si, Ch., Be uccio, P., Mo andi, A., Le i, F., Bose i, C. and C. La Vecchia (2012): ‘T ends in mo o ehicle c ash mo ali y in Eu ope, 1980-2007’, Sa e y Science, 50, 1009-1018. Palk, G.R.M. and J.R. Da ey (2005): A compa a i e analysis o he na u e and ex en o a ic o ences and hei ela ionship o o he non a ic o ences. The Road Sa e y Resea ch, Policing and Educa ion Con e ence P oceedings, Pe h, Aus alia. Pe idou, E. and M. Mous aki (2000): ‘Human ac o s in he causa ion o oad a ic c ashes’. Eu opean Jou nal o Epidemiology, 16, 819-826. Pe ie, D. J., Do an, Ch. M., Shakesha , A. P. and R. Sanson-Fishe (2010): ‘The ela ionship be ween isky alcohol consump ion, c ime and a ic acciden s in Aus alian u al communi ies’, Addic i e Beha io s, 35, 359-362. Po e ield, A.L. (1960): ‘T a ic a ali ies, suicide and homicide’, Ame ican Sociological Re iew, 25, 897-901. Rose, G. (2000): The c iminal his o ies o se ious a ic o ende s. Home O ice Resea ch, De elopmen and S a is ics Di ec o a e. Resea ch Findings No. 206, London. Sachsida, A., Ca doso de Mendonça, M. J., Lou ei o, P.R.A. and M.B. Sa mien o Gu ie ez (2010): ‘Inequali y and c iminali y e isi ed: u he e idence om B azil’, Empi ical Economics, 39, 93-109. Săucan, D.-S., Micle, M.-I., Popa, C. and G. Oancea (2012): ‘Violence and agg essi eness in a ic’, P ocedia –Social and Beha io al Sciences, 33, 343-347. Sen, A. (2001): ‘Do s ic e penal ies de e d inking and d i ing? An empi ical in es iga ion o Canadian impai ed d i ing laws’, Canadian Jou nal o Economics, 34, 149–164. 23 Si ak, M. (1983): ‘Socie y’s agg ession le el as a p edic o o a ic a ali y a e’, Jou nal o Sa e y Resea ch, 14, 93-99. Smink, B. E., Rui e , B., Lus ho , K. J., de Gie , J. J., Uges, D.R.A. and A.C.G. Egbe s (2005): ‘D ug use and he se e i y o a a ic acciden ’, Acciden Analysis and P e en ion, 37, 427-433. S anoje ić, P., Jo ano ić, D. and T. Lajunen (2012): ‘In luence o a ic en o cemen on he a i udes and beha io o d i e s’, Acciden Analysis and P e en ion, 52, 29-38. Süme , N. (2003): ‘Pe sonali y and beha io al p edic o s o a ic acciden s: es ing a con ex ual media ed model’, Acciden Analysis and P e en ion, 35, 949-964. Tay, R. (2005): ‘Gene al and speci ic de e ence e ec s o a ic en o cemen . Do we ha e o ca ch o ende s o educe c ashes?’, Jou nal o T anspo Economics and Policy, 39, 209- 223. Tay, R. (2010): ‘Speed Came as Imp o ing Sa e y o Raising Re enue?’, Jou nal o T anspo Economics and Policy, 44, 247-257. Ve eeck, L. and K. V olix (2007): ‘The social willingness o comply wi h he law: The e ec o social a i udes on a ic a ali ies’, In e na ional Re iew o Law and Economics, 27, 385-408. Wegman, F. C. M. (1992): Legisla ion, egula ion and en o cemen o imp o e oad sa e y in de eloping coun ies. Con ibu ion o he Wo ld Bank Semina on Road Sa e y, Washing on. Wes , R., Elande , J. and D. F ench (1993): ‘Mild social de iance, Type-A beha iou pa e n and decision-making s yle as p edic o s o sel - epo ed d i ing s yle and a ic acciden isk’, B i ish Jou nal o Psychology, 84, 207-219. 24 Wilson, J. and G. Kelling (1982): ‘B oken windows: he police and neighbo hood sa e y’, A lan ic Mon ly, 256, 29-38. Zaal, D. (1994): T a ic law en o cemen : A e iew o he li e a u e. Monash Uni e si y Acciden Resea ch Cen e. Repo Documen No. 53. Aus alia. TABLE 1. Va iables used in he empi ical analysis. Va iables Desc ip ion Sou ce Mean S anda d De ia ion Pe capi a a ali ies Fa ali y a es pe million inhabi an s CARE (EU oad acciden s da abase) 108.44 45.04 Li e imp isonmen Index a iable ha akes he alue 0 i no li e imp isonmen sen ence may applied; 1 i imp isonmen sen ence is e isable (pa don, amnes y o o he elease may be applied); 2 i inde ini e sen ence is no e isable INTERNATIONAL JUSTICE RESOURCE CENTER 1.57 0.67 Homicides Numbe o homicides pe 100,000 inhabi an s (In en ional killing o a pe son, including mu de , manslaugh e , eu hanasia and in an icide) EUROSTAT 2.24 2.31 Robbe y Robbe y c imes pe 100,000 inhabi an s (S ealing om a pe son wi h o ce o h ea o o ce, including muggings- bag-sna ching and he wi h iolence) EUROSTAT 88.83 70.35 25 D ugs d ug a icking c imes pe 100,000 inhabi an s (Illegal possession, cul i a ion, p oduc ion, supplying, anspo a ion, impo ing, expo ing, inancing e c. o d ug ope a ions which a e no solely in connec ion wi h pe sonal use) EUROSTAT 66.34 65.82 Mo o ehicle he Mo o ehicle he pe 100 000 inhabi an s (The o all land ehicles wi h an engine ha un on he oad which a e used o ca y people, including ca s, mo o cycles, buses, lo ies) EUROSTAT 219.55 184.68 Pe capi a GDP Pe capi a g oss domes ic p oduc in In e na ional Compa able P ices (US$ a 2005 p ices and PPP) UNECE S a is ical Di ision Da abase, compiled om o icial na ional and in e na ional (CIS, EUROSTAT, IMF, OECD) sou ces 24960 11653 Mo o iza ion Numbe o egis e ed ehicles pe 1000 inhabi an s UNECE anspo di ision, Eu os a , Wo ld Bank and na ional da abases 424.74 111.99 Vehicles-km d i en Numbe o passenge -ca s-km exp essed in 1000 million km and weigh ed by na ional popula ion Eu opean Commission (Di ec o a e Gene al o mobili y and anspo ) 39.97 317.52 Mo o way densi y Numbe kms o mo o ways di ided by squa e km o he coun y EUROSTAT and UNECE 1.69 1.73 Old % popula ion o e 60 yea s old EUROSTAT 20.67 2.52 Young % popula ion aged 20-39 yea s EUROSTAT 28.75 1.92