International Journal of Police Science & Management Volume 14 Number 3 Explaining the public distrust of police in the newest European Union countries Viviana Andreescu† and Deborah G. Keeling‡ †(Corresponding author) Department of Justice Administration, University of Louisville, Louisville, KY 40292, USA. Email: v0andr01@louisville.edu ‡Department of Justice Administration, University of Louisville, Louisville, KY 40292, USA Email: dgwils01@gwise.louisville.edu Submitted 1 October 2011, revision submitted 9 March 2012, accepted 29 March 2012 Keywords: trust, trust in police, police reform, Bulgarian police, Romanian police Viviana Andreescu is an assistant professor in the Department of Justice Administration at the University of Louisville. She holds PhD, MA and MS degrees from the University of Louisville and a Bachelor of Arts in sociology from the University of Bucharest, Romania. Deborah Keeling is professor and department chair. She oversees academic programmes in the Department of Justice Administration, the Southern Police Institute and the National Crime Prevention Institute at the University of Louisville. She holds a PhD in sociology from Purdue University. the police of subjective (eg, perceptions of important legal and political institutions; interpersonal trust; perceived sense of safety; perceived social and economic exclusion) and objective individuallevel factors, such as sociodemographic characteristics, residency in capital cities (Bucharest and Sofia) and experiences with victimisation. Results show that, in both countries, institutional trust is the most important predictor of public attitudes toward the police. The potential impact of real and perceived recent crime trends and the latent effects of economic, political, legal and historical conditions on public discontent with the police in developing democracies is also discussed. ABSTRACT Based on recent cross-sectional data from the European Social Surveys Round 4 (ESS4), in Romania and Bulgaria, the average levels of confidence in the country’s police appear to be much lower than in most European Union states. However, Romanians and Bulgarians tend to trust the local police more than they trust their country’s legal system, the main political parties, the country’s parliament and the national government. Using ESS4 data collected from national representative samples of Romanians and Bulgarians, the present quantitative analysis attempts to assess the relative influence on attitudes toward INTRODUCTION Approximately 30 years ago, police practitioners and scholars began advocating a change in police values and started to acknowledge the role of citizens in the production of justice, which is now considered a founding principle of modern policing (Dowler & Sparks, 2008). In this context, it became important to identify people’s expectations and perceptions of policing. The police can be considered the most visible agents of governmental authority and power, and what people think about International Journal of Police Science and Management, DOI: 10.1350/ijps.2012.14.3.278 Page 1 Explaining the public distrust of police in the newest EU countries the police and their work should be known because it could serve as an important social indicator of the political health of a society as a whole (Benson, 1981). Knowing residents’ attitudes toward the police is equally important for law enforcement agencies. As one scholar noted, ‘popular opinion matters in part because widespread confidence in the police makes law enforcement officers’ work easier and more effective’ (Skogan, 2006, p. 118). Institutional legitimacy of the police through public trust is particularly needed in democratic societies because it may strengthen cooperation between citizens and the criminal justice system. By making people believe that their long-term personal interests will be well served, public voluntary compliance with the law and public commitment to the rule of law may increase (Hinton & Newborn, 2009; Sztompka, 1998). As Jackson and Bradford (2009) observed, legitimacy encourages people to follow the rules, not out of fear of punishment, but because they believe they ought to. In sum, ‘the criminal justice system relies on legitimacy and consent to an extent unlike other public services: public support is vital if the police and other criminal justice agencies are to function both effectively and in accordance with democratic norms’ (Jackson & Bradford, p. 493). Although a large amount of research on public perceptions of the police has been conducted in the USA (Ackerman et al., 2001; Benson, 1981; Dowler & Sparks, 2008; Frank, Smith, & Novak, 2005; Hurst, 2007; Maxson, Hennigan, & Sloane, 2003; Reisig & Parks, 2000; Scaglion & Condon, 1980; Weitzer & Tuch, 2005), studies on attitudes towards the police have been carried out in Europe also. In Western Europe, for instance, the relationship between the police and the public has been studied extensively in the UK (Bradford & Jackson, 2011; Bradford, Jackson, & Stanko, 2009; Page 2 Hough, Jackson, Bradford, Myhill, & Quinton, 2010; Jackson & Bradford, 2009; Jackson, Bradford, Hohl, & Farrall, 2009; Jackson & Sunshine, 2007; Myhill & Quinton, 2010). With regard to the states of the former communist bloc, Russia, more than any other country in the region, has attracted the attention of a relatively large number of researchers interested in residents’ satisfaction with the police or the public’s experiences with the police (cf. Reynolds, Semukhina, & Demidov, 2008). Yet a large part of the research into Europeans’ attitudes toward the police has used aggregate-level data and analysed public perceptions of the police by comparing two entities — Western European countries vs. Central and Eastern European countries (see Aromaa, Leppa, Nevala, & Ollus, 2003; Kaariainen, 2007; Mueller, 2008). Overall, findings based on analyses of surveys conducted in Europe consistently showed that, in the former communist states, average levels of public satisfaction with the police were significantly lower than in Western European democracies (Kaariainen; Zvekic, 1996). Despite the fact that police reform has been widely undertaken in postauthoritarian states, such as Romania and Bulgaria, the absence of public trust in the police continues to characterise police– community relations in these countries (see Goldsmith, 2005). Therefore, explaining the nature of police mistrust becomes an important starting point in the process of building public confidence in a vital institution of the state. This issue is particularly relevant now, when both countries are trying to meet European Union (EU) standards of modern policing. Although our secondary analysis has been constrained by the relatively limited information provided by ESS4 survey data (European Social Survey [ESS], 2008), we tried to identify some of the personal experiences and background variables that Andreescu and Keeling might play a role in structuring attitudes towards the police in Romania and Bulgaria. To the authors’ knowledge, the present research is the first comparative multivariate analysis that focuses exclusively on Romanians’ and Bulgarians’ current opinions about the police. The main objectives of our research are: (1) to examine the nature of trust in the police in general terms and in relation to Romanian and Bulgarian citizens’ trust in governance; (2) to identify the factors most likely to predict the public’s distrust of the police in Bulgaria and Romania. Both countries share a communist past, and they both joined the EU most recently, on 1 January 2007. POLICING IN BULGARIA AND ROMANIA Brief historical overview The institution of police was officially created in Bulgaria in the 19th century, with the adoption of the Constitution of Tarnovo on 16 April 1879. By Decree No. 1 of 5 July 1879, issued by Prince Alexander 1st, the Ministry of Internal Affairs was set up as part of the Bulgarian government (Ministry of Interior [MoI], n.d.). In 1806, those in charge of crime control and maintaining order in Bucharest, the capital of Wallachia, were referred to as the police for the first time (Politia Romana, n.d.). Ministries of Domestic Affairs were created in 1831 and 1832 in Wallachia and Moldavia, respectively, when quasi-constitutional organic laws (ie, the Organic Regulations) were enforced by the Imperial Russian authorities in the two Principalities that were to become the basis of the modern Romanian state. In 1862, Alexandru Ioan Cuza, the ruler of the United Romanian Principalities signed Decree No. 495 and the Ministry of Interior, with its headquarters in Bucharest, was formed (Ministry of Administration and Interior [MAI], n.d.). Reorganisation of the police continued throughout the 19th century and in 1850 the first centralised structure in charge of police organisation and coordination was formed. From 1860 to 1929, three new laws referring to the organisation, structure and functions of the police were enacted. After the Second World War, Decree No. 25 of 1949 stipulated the creation of the militia as the main law enforcement agency in the country. Forty years later, the name of the institution was changed from militia to police when Law No. 2/ 27/12/1989 (regarding the constitution, organisation and functioning of the Council of National Defence) was enacted (Abraham, 2001; Politia Romana). During communist times, the main task of the Bulgarian and Romanian police forces (ie, militia) was to sustain the political system and protect its functioning. Similar to law enforcement in other Eastern European countries, the police operated as a subsidiary of the state security agency and were under the control of the Interior Ministry. Regular police work, such as crime prevention and control was secondary to the task of maintaining the security of the state (Koszeg, 2001). In addition, several former communist regimes sought to combine the military and the police in a suprainstitution of state coercion that could be employed against any internal threats to the party leadership (Watts, 2004). In Romania, the police command was completely politicised, and all police leaders were members of the Communist Party. Police recruits were not selected based on their professional aptitudes and integrity. Having a ‘healthy’ social class origin (ie, peasant or working-class family background) was a decisive factor in recruitment to the police force (Abraham, 2001). Prior to 1989, the screening, appointment, promotion and dismissal of police officers were based on political considerations in Bulgaria as well. Legal regulations regarding personnel matters were obscure, and job security was Page 3 Explaining the public distrust of police in the newest EU countries mostly a matter of obedience (Bojkov, 2006). Police reform and challenges of the transition toward democratic policing Since the beginning of the 1990s, the Eastern European police forces have gone through a process of reform (see Bojkov, 2006; Caparini & Marenin, 2005; Hinton & Newborn, 2009; Kadar, 2001; Koci, 1996; Mobekk, 2005; Uildriks & van Reenen 2003). These reforms focused on institutional restructuring and legal redress. A main objective was to integrate police services into a system of law application consistent with the accepted norms and practices in liberal democratic states (Bojkov). However, as Hinton and Newborn noted, ‘reform is a process of shifting sands as old habits and cultural patterns die hard and traditional practices often reemerge under a different guise’ (p. 23). A Bulgarian scholar (Bojkov, 2006) acknowledged that reform of the police and its specialised services has been relatively slow in Bulgaria. According to Bojkov (2006), the most relevant reasons for this were the political standing and functions of the police prior to 1989, the complex institutional set-up that made the police dependent on other state structures, and the lengthy process of Bulgarian police integration into international organisations that encouraged transparency and public accountability. Mobekk (2005) also noted that the Bulgarian police did not benefit from large support operations and that, in general, the international community has been less involved with policing, police assistance and police reform in Bulgaria compared with other former communist countries. Although accountability structures have been established with the support of the EU, the use of these structures was not found to be effective. Police reform in Bulgaria has also been negatively affected by an ineffective judicial system (Mobekk). Page 4 Since the beginning of the reform in 1990, several new institutions have been created in Bulgaria only to be amended or abolished after a short period. In particular, since 1990, the Code of Criminal Procedure (CCP), which was initially adopted in 1974, has been subjected to a vast number of amendments in response to radical changes in the Bulgarian political, economic and legal systems. However, ongoing amendments to the CCP were sometimes contradictory and often clashed with other legislation, ‘making criminal law enforcement extremely difficult and inefficient’, as one Bulgarian scholar noted (Marinova, 2006, p. 77). In Bulgaria, the MoI has sole authority over the agencies acting within its framework. The main structural units are: the National Security Service, the National Police Service, and the National Fire Safety and Protection of Population Service. The National Police Service includes regional directorates of police and the General Directorate of Police with eight agencies (eg, Chief Directorate for Combating Crime, Protection of Public Order, and Criminal Prevention; Chief Directorate of Combating Organised Crime; Chief Directorate Border Police; Chief Directorate Gendarmerie; Migration Directorate; International Operational Police Cooperation Directorate; Antiterrorism Task Force; Aviation Task Force) (MoI, n.d.; Organization for Security and Co-operation in Europe [OSCE], 2009). Bulgarian jurisprudence regards police employees as civil servants. However, the legal status of police officers is different from that of ordinary civil servants with respect to employment, dismissal and promotion (Weber, 2001). Although there were no provisions in the 1991 Constitution guaranteeing a clear separation between the army and the police, the army does not have law enforcement powers in Bulgaria, with the exception of the military Andreescu and Keeling police. Until recently, as a consequence of the quasi-military character of the Bulgarian police, several aspects related to police organisation and structure were considered ‘state secrets’ (Weber). For example, police numbers were made public for the first time only in January 2010, when the Bulgarian Minister of Interior announced at a press conference that the MoI has 61,170 employees and that 60 per cent of them (36,727) are police officers, distributed in 34 directorates. According to the press release, approximately 75 per cent of police officers work in 28 regional directorates; 16 per cent belong to the main directorate of border police; 5 per cent belong to the main directorate of security police; 3 per cent are employed by the main directorate of criminal police; and fewer than 1 per cent of the police officers work in the international operational police cooperation directorate, in the main directorate for pretrial proceedings, and in the internal security directorate. Compared with other EU states, Bulgaria has a large number of police officers relative to its population size. Based on 2010 data, the average number of police officers in EU was 339 per 100,000 people, whereas there were 481 police officers per 100,000 people in Bulgaria. Estimates show that in 2010 there were 211 police officers per 100,000 people in Romania (Slavova, 2010). In August 2011, the Bulgarian MoI announced that 28 measures would be implemented in response to the recommendations and criticisms formulated by the European Commission under the mechanism for verification of Bulgaria’s progress in justice and home affairs. Among the top priorities were listed: continuation of criminal police reform, more serious investigation into sophisticated economic and financial crimes, improvement in the capacity of investigative bodies in connection with the launch of the new specialised court of justice, and interdepartment integration. Reform of the criminal police legislation also includes changes to the Criminal Procedure Code and the Law on the Ministry of Interior. Until 2010, the Criminal Procedure Code stipulated that only 2,000 officials from the MoI could carry out investigations, which amounted to 200,000 cases annually in Bulgaria. Uniformed and ordinary police officers were not allowed to conduct crime scene investigations, to interrogate witnesses or interview victims of crime. As a result of the new legislative changes, some 4,000 police officers have joined the team of MoI investigators. In order to avoid overlaps and increase efficiency, the directorates of the Interior Ministry have been reduced in number (Tsvetkova, 2011). It remains to be seen whether these legislative and institutional changes will have the anticipated positive effects. In the early 1990s, in Romania, there was substantial internal and external pressure to reform major societal institutions, including the police. An important step was to depoliticise police structures and to eradicate the political bureaus that existed within the police and security sector. Consequently, in the first four years of the reform, approximately two thirds of police managers were replaced. However, these changes seriously undermined the professionalism of the police force, because the new leaders appeared to be less qualified than their predecessors. Moreover, in 2000, many police commanders were in leadership positions not because they were highly competent, but mostly as a result of the informal connections they had with the new political elite (Abraham, 2001). Even if initially there was quite substantial opposition to police reform in Romania, post-1996, the police and civil society were the key actors pushing for transformation and change (Mobekk, 2005). Nevertheless, a former high-rank Page 5 Explaining the public distrust of police in the newest EU countries police official noted that, from 1990 to 1999, neither a definite reform plan nor a coherent strategy has been elaborated in Romania. Things started to change in the late 1990s when cooperation agreements were established with specialised institutions from the USA, Italy, Germany, France and the UK. In addition, foreign specialists collaborated with the Romanian Ministry of Interior and offered useful advice regarding police reform (Abraham, 2001). However, according to an analyst of police reform in south-east Europe, Romania received less international assistance and support for police reform than other countries in the region. Yet it should be noted that the Foreign and Commonwealth Office in UK provided training and equipment in specialised police work and accountability, and organised courses focusing on the fight against organised crime for elite officers. The Council of Europe has supplied expert assistance for community policing and, in 2003, organised a course on ethical police behaviour. In addition, the Swiss Agency for Development and Cooperation (SDC) organised training sessions, seminars and workshops focusing on the prevention of economic crime and money laundering (Mobekk). In Romania, for more than one decade after the fall of Ceausescu’s totalitarian regime, the legal status of police officers was similar to that of military personnel. There were ranks and rules concerning promotion, discipline and subordination to superiors. Police officers had several privileges including higher than average salaries, bonuses, free uniforms and civilian clothing, and free medical care in hospitals belonging to the MoI. Romanian police officers were not allowed to join political parties, go on strike or set up their own trade unions or professional associations. If dismissed, police officers had no resort to a judicial review (Weber, 2001). The status of the police changed in June 2002 (Law Page 6 360/2002), when the Romanian police was demilitarised. The most recent legislative changes regarding the police were included in the Law 133 from June 2011 (Politia Romana, n.d.). All Romanian police officers have an obligation to respect the rules and regulations stipulated in the Romanian police officer’s Code of Ethics and Deontology. The code of ethics acknowledges that a police officer has to show integrity (ie, to behave in accordance with the ethical norms accepted and practised in society) and has to demonstrate loyalty toward the police and its institutional values. Article 19 of the code specifies that a police officer is not allowed to tolerate corruption and abuse the public authority invested in the police. The code also forbids a police officer to demand or accept money or other material gains as compensation for fulfilling his/her professional duties. According to the code, a police officer should take action against any internal institutional corruption, having the obligation to inform his or her superiors and other specialised units about any act of corruption the officer becomes aware of. Regarding the use of force by police, Article 9 of the code states that the police should use excessive force only in extreme situations, in response to serious threats or physical violence directed at them or other persons. The code also stipulates that the police officer should always show respect for the human dignity when making use of force in the execution of police duties (Guvernul Romaniei, 2005). Currently, the Romanian police is the national police force and the main civil law enforcement agency in Romania (OSCE, 2010). A subordinate agency of the MAI, the Romanian police is responsible for: the protection of fundamental rights and freedoms, the protection of private and public property, the prevention and identification of criminal offences and their perpetrators, and maintaining public order (Law Andreescu and Keeling 218/2002, Art. 1). Romanian police headquarters — known as the ‘General Inspectorate’ — coordinates police action across the country. It investigates major cases related to serious, organised and financial crime. The General Inspectorate is under the command of an Inspector General appointed by the MAI. In addition to nine distinct units (eg, Directorate for Operations Management, Directorate for Budget and Accounting, Directorate for Internal Control, and Directorate for Intelligence Management and Analysis) there are four General Directorates (ie, Public Safety, Criminal Investigations, Countering Organised Crime, and Administrative Police). There are also 41 territorial police units (Politia Romana, n.d.). In 1990, there were 32,748 police officers in Romania (Abraham, 2001) whereas, according to the 2010 report of the MAI, in 2010 54,791 persons were employed by the Romanian police. The Romanian border police and the Romanian gendarmerie are distinct units also subordinated to the MAI (MAI, 2011). The Romanian gendarmerie is a specialised institution with military status, which has the following attributions: protection of public order and safety, protection of citizens’ rights and fundamental liberties, protection of public and private property, crime prevention and detection, protection of the state’s fundamental institutions, and countering terrorist acts (MAI, n.d.). In 2010, the Romanian gendarmerie had 28,515 employees and the Border Police had 16,430 employees (MAI, 2011). In common with Bulgaria, reform of the Romanian police has been a slow process. A relatively high cost of reforming police structures, a pervasive fear that a civilian police force would be less efficient and persistence among the police of certain mentalities rooted in the past delayed the reform (Abraham, 2001). Frequent legislative changes and numerous changes in leadership (eg, in the 21 years since December 1989, the MAI minister has been changed 17 times) also negatively affected the pace of police reform in Romania. An examination of recent national trends regarding the judicial, administrative and anti-corruption reforms in Romania showed that, from 1999 to 2007, backtracking (ie, amendments that dilute the reforms, annulment of government ordinances by parliament or any other measures with the purpose of diminishing the capacity and scope of reforms) occurred in 49 of the 61 cases of legislative adoption or institutional creation (Racovita, 2011). Although significant structural changes in law enforcement did eventually occur in Romania (eg, demilitarisation of the police, and the creation of new specialised units to fight organised crime, illegal immigration, human trafficking and corruption), and efforts to increase police transparency and accountability can be noticed (eg, since 2005, annual performance evaluation reports have been available on the MAI Website), an integrated long-term strategy regarding police reform is still missing. In recent years, many institutional reforms in Romania and Bulgaria have been, in part, the result of strong international pressure. Nonetheless, since 2006, when the European Commission Decision 2006/928/EC established a mechanism for the verification of progress in Romania, the focus has been on judicial reform and fight against corruption (see European Commission, 2011, 2012a) and less on police reform. It can be argued that the concentration of efforts on reform of the judicial system has affected the crystallisation of police reform in Romania. For instance, the most recent report of the European Commission on progress in Bulgaria under the cooperation and verification mechanism states that, in addition to continuation of the reform of the judiciary and the fight against corruption, the country should continue reform of Page 7 Explaining the public distrust of police in the newest EU countries the police and improve the effectiveness of the fight against organised crime (European Commission, 2012b). No references to police reform are made in the corresponding report of the European Commission on progress in Romania (see European Commission, 2012a). Police–public relationships In a review of police reforms in postsocialist Central and Eastern European states, Caparini and Marenin (2005) concluded that, despite structural and procedural reforms implemented in police organisations across the region, overall results are not satisfactory, especially when the views of citizens and the victims of crime are considered. In the view of Caparini and Marenin, the continuing low rate of public confidence in the police, citizens’ reluctance to report crime and the high levels of fear of crime indicate that the relationship between the police and local communities should change and that police performance in crime prevention and control needs to improve significantly. Hinton and Newborn (2009) noted that scandals concerning overpolicing, underprotection, racial and ethnic profiling, lack of respect for legal procedures, capitulation to political pressure, and arbitrary and abusive practices are far from uncommon in developing countries. Some of these issues have been identified in Bulgaria and Romania when policing in these countries has been examined. Even though several police officers trained prior to 1989 left the force in the early 1990s (see Abraham, 2001; Mobekk, 2005), in 2000, a considerable number of serving police officers in Bulgaria and Romania were persons selected and trained under the communist system. Many of these ‘old-new’ police officers found it difficult to erase memories of the inheritance of the past, especially with regard to the role of the police in society (Rzeplinski, 2001). In addition, like most Page 8 countries in the region, after the fall of the communist regimes, Bulgaria and Romania experienced significant increases in crime rates, which generated an enormous pressure on police to achieve results. Since 1990, Amnesty International has consistently recorded cases of police brutality, ill-treatment of alleged perpetrators, and the extensive use of firearms by the Bulgarian police (Mobekk, 2005). A survey conducted among Bulgarian police officers in 2000 by the World Organisation Against Torture (OMCT — Europe) indicated that officers felt that they could use force to control the population and as a crime prevention tool. More than one third of the officers interviewed at the time admitted to committing acts that would constitute an illegal or unethical use of force (Mobekk). Novinite, the Sofia news agency, recently reported that in the 12 years to 2010, the European Court of Human Rights had delivered 27 guilty verdicts against Bulgaria concerning the excessive use of force by police officers (Novinite, 2012). In addition to these documented cases of abuse of power by the police, a perceived increase in organised crime, the feeling that law enforcement works exclusively for the politically powerful and perceptions of the spread institutional corruption have contributed to an increased sense of insecurity among the population and a diminished level of confidence in the Bulgarian police (Spirova, 2009). Mishler and Rose (1997) noted that even if in the short term, institutional trust may be an inherited legacy of a communist regime, in the longer term, trust must be earned and must be performance based. At the end of 2010, fewer than half of Bulgarian citizens made a positive evaluation of the police performance. Despite the fact that recent data indicate that the number of registered and unreported crimes has Andreescu and Keeling declined, the popular perception in Bulgaria is that crime is on the increase and the authorities in charge of crime prevention and control are largely inefficient. Bulgarian analysts consider that an overexposure of crime in the media and the fact that crime prevention measures have become a basic political and pre-election tool, have affected citizens’ sense of safety and were some of the main reasons for the negative popular assessment of police work. In addition, the low level of public confidence in the police is considered to be a reflection of a generalised low level of trust in society, at both interpersonal and institutional levels (Center for the Study of Democracy, 2011). Hinton and Newborn (2009) contended that police involvement in repression associated with authoritarian regimes has cast a long shadow over police–community relations, generating enduring patterns of mutual suspicion, fear and mistrust. In the early 1990s, in particular, it can be argued that this was the case in Romania. During the events of December 1989, public anger toward the police was evidenced by attacks directed by the population at several police stations, which were set on fire or destroyed. In these unprecedented violent circumstances, 42 police officers were killed and 60 were seriously wounded. Public discontent with the Romanian police was reinforced in the following two years, when during the miners’ marches of June 1990 and September 1991, the police were perceived as supporters of the miners’ violence and of a political regime contested by a significant segment of the population, particularly in Bucharest, the capital city of Romania. Police officers’ inadequate and abusive interventions in disputes involving ethnic minorities, such as the Roma minority, have also contributed to the negative image of Romanian law enforcement (Abraham, 2001). Data from the New Democracy Barometer (NDB), based on surveys conducted on representative samples between November 1993 and March 1994 in seven Central and Eastern European countries, show that, on a scale from 1 (great distrust) to 7 (great trust), the average level of trust toward the police in Romania was 3.8, slightly below the middle of the interval and lower than the level of public trust in the Romanian army (5.5). It can be said that scepticism characterised Romanians’ attitudes toward the local police at the time. Based on NDB results, in Bulgaria, the average levels of public trust in the police (2.9) and in the military forces (4.6) were lower than in Romania (Mishler & Rose, 1997). In 2000, the Romanian police were frequently criticised for their insufficient organisational adaptation to rapid societal changes. Charges that the police continued to serve party interests, incidents of police brutality and police corruption reported by the media, and a significant increase in crime levels contributed to the erosion of public trust in the police (Abraham, 2001). This trend continued in the following decade. Based on data collected by the most recent Global Corruption Barometer, the largest cross-country survey of the general public’s views on and experiences of corruption, on a scale from 1 (not at all corrupt) to 5 (extremely corrupt), the average score for perceived police corruption obtained in Romania was 3.9; the highest value in the EU. The average score for perceived police corruption in Bulgaria was slightly lower (3.8) (Transparency International, 2010). A performance evaluation report of the Romanian MAI showed that, in 2010, of 1,503 persons accused of corruption by the National Anticorruption Directorate (DNA), 222 were police officers or police agents. Although the police officers/agents (N = 109) found guilty of corruption or serious police misconduct represent only Page 9 Explaining the public distrust of police in the newest EU countries 0.20 per cent of the total number of persons employed by the Romanian police (N = 54,791), it should be noted that, compared with previous years, the data show an increase in the number of cases of corruption that involved the Romanian police (MAI, 2011). Despite an overall positive evaluation of police activity in 2010, and significant decreases in the levels of violent crime, human trafficking and cybercrime (see MAI, 2011), recent research shows that Romanians continue to express a low level of confidence in the police. According to the results of a public opinion poll conducted in 2010 on a representative sample, approximately 66 per cent of Romanians indicated that they do not trust the police (Zavaleanu, 2010). Similar to Bulgarian scholars, Romanian analysts believe that the low level of public trust in the police is a consequence of a generally low level of public confidence in major institutions and a side effect of the lack of credibility associated with politicians and the main political parties. These beliefs are considered to be a result of perceived widespread corruption and increases in poverty rates (see Deleanu & Bechir, 2011). In his analysis of data collected in Romania, Uslaner found strong links between economic inequality, perceptions of corruption and social trust. According to Uslaner (2010), corruption persists in societies such as Romania, where an inequality trap can be identified. Describing the essence of the ‘inequality trap’, Uslaner argues that ‘the roots of corruption rest upon economic inequality and low trust in people who are different from yourself ’. Corruption, in turn, leads to more inequality and to less interpersonal trust. And, as many studies have shown, low interpersonal trust is associated with low institutional trust (cf. Rothstein & Uslaner, 2005). Page 10 ABOUT TRUST AND ITS SOCIAL RELEVANCE Over the past two decades, issues related to trust have been covered extensively in the social science literature. According to Rothstein and Uslaner (2005), from 1990 to 2005, the number of articles published in scholarly journals that listed ‘trust’ as a keyword, increased from 129 to 1,956. The British philosopher Onora O’Neill noted that ‘trust is constantly observed, is hard earned and easily dissipated. It is valuable social capital and not to be squandered’ (O’Neill, 2002, p. 6). As Sztompka (1999, p. 5) emphasised, in addition to loyalty and solidarity, trust, briefly defined as ‘the expectancy of others’ virtuous conduct toward ourselves’, is one of the basic components of a moral community. Sztompka (1999) argues that social trust is particularly important in contemporary societies. One reason is that we moved from societies based on fate to those advanced by human agency, and people need to deploy trust in order to face the future actively and constructively. In addition, the ongoing process of global interdependency will increase the demand for trust as an essential condition for cooperation. Trust is particularly needed because we live in a ‘risk society’ that makes humans more vulnerable to the expanding threats and hazards that pervade our social life. Trust also becomes an important decision factor when people face so many options and alternatives. The complexity of institutions, organisations and technological systems, and the increasingly global scope of their operations make our social environment increasingly opaque. Therefore, trust is needed for humans to be able to act in the current social environment that many people perceive as impenetrable. In addition, trust is essential in contemporary societies because humans have to rely more and more on anonymous ‘significant others’ whose actions constantly Andreescu and Keeling impact people’s wellbeing. Finally, considering the fact that contemporary societies are characterised by demographic diversity generated by recent intense migratory flows, social trust becomes a necessary resource when an increasing number of people have to cope with strangers (Sztompka). As Uslaner (2002) observed, trust lays the basis for cooperation with people who are different from yourself. Barbalet (2009) noted that trust is a type of belief or faith that can never be based on pertinent knowledge or rational calculation; it has a creative capacity and represents an anticipation of a future positive outcome that could not occur without the giving of trust. Although trust faces uncertainty, a sense of certainty can be achieved through organisation, contract, sanctions, and incentives. These, however, cannot be the bases of trust. ‘There is essentially an emotional basis that is the foundation to all trust’ (Barbalet, pp. 368–377). Discussing the varieties of trust, Uslaner (2005) made a distinction between ‘strategic trust’ (ie, the standard meaning of trust that reflects our expectations about how people will behave) and ‘moralistic trust’ (ie, a statement about how people should behave). Moralistic trust does not stem from experience as strategic trust does. Moralistic trust is rather stable over time, is learned early in life and is conducive to civic engagement (eg, leads people to contribute to charity, to volunteer and to be more tolerant of minorities). According to Uslaner (2002, 2005), moralistic trust creates the foundation for ‘generalised trust’, which is the faith in others or the perception that most people are part of your moral community. Uslaner (2002) differentiates between ‘generalised trusters’ (persons who believe that most people share common values and are willing to trust strangers who may outwardly seem quite different from themselves) and ‘particularised trusters’ (persons who do have faith in other people but only in other people from their own group) and states that there is a continuum from particularised to generalised trust (Uslaner, 2005). According to the author, trust benefits are a result of generalised trust (Uslaner, 2002). Regarding institutional trust, Devos, Spin, and Schwartz (2002) acknowledged that trusting an institution entails having confidence that the institution is reliable, observes rules and regulations, works well and serves the general interest. ‘If someone trusts an institution, it implies that he or she believes that this collective entity, on the whole, is competent, fulfils its obligations, and acts in responsible way’ (p. 484; cf. Kaariainen, 2008). In sum, as Barbalet (2009) pointed out, ‘when citizens are asked if they trust certain institutions the question refers to the legitimacy or the performance of the institutions’ (p. 369). In addition, as Kaariainen observed, public trust toward public services (eg, the police, the judicial system) that should be available to all citizens alike imply a public expectation of impartiality. From the 1960s to 1990s, a sharp and linear decline in trust could be observed in the USA (Uslaner 2002). Referring primarily to the British society, O’Neill (2002, p. 6) observed that the ‘loss of trust appears to be a cliché of our times’. However, because conclusive evidence that people or institutions are less trustworthy is elusive, the author considers that the supposed ‘crisis of trust’ indicates in fact the existence of a culture of suspicion (O’Neill). Although it might be true that in certain developed societies, such as the UK or USA, the levels of social trust decreased over the past decades, higher levels of interpersonal and institutional trust continue to be detected in industrialised societies compared with other countries (cf. Rothstein & Uslaner, 2005). Uslaner (2002) observed that most trusting countries (without a legacy of communist rule) are less corrupt, have better judicial systems, less red tape in Page 11 Explaining the public distrust of police in the newest EU countries bureaucracies, greater government spending (especially in education) as a percentage of gross domestic product, more redistribution of wealth from the rich to the poor and more open economies. Sztompka (1998) noted that the factors that determine variations in levels of interpersonal and institutional are: the ‘reflected trustworthiness’ of the target (social agent or organisation), which is estimated by individuals in a more or less rational manner; the attitude of ‘basic trustfulness’ deriving from socialisation; and the ‘culture of trust’ pervading the society and normatively encouraging the trusting orientation. In Uslaner’s view, ‘a culture of trust depends upon the idea that things will get better for those who have less and that it is in our power to make the world better’ (p. 3). In other words, a culture of trust may be found in countries characterised by economic equality and equality of opportunity, two different, yet inter-related types of equality that are considered by Rothstein and Uslaner (2005) the main determinants of trust. As documented by research, most former communist countries are among the least trusting societies (Rothstein & Uslaner, 2005). Sztompka (1999) considered that a ‘culture of mistrust’ was reinforced in these countries by certain characteristics of the political system, such as the high levels of uncertainty that citizens faced during the adoption of the new democratic system, an inefficient monitoring of institutions that had to guarantee law and order, the image of the new political elites as self-interested and higher expectations raised in the years of the transition. Following Uslaner (2002, 2010), we can argue that increasing economic inequality and widespread corruption in countries such as Romania and Bulgaria also contributed to persistent low levels of institutional trust in Eastern Europe. The fact that the police reflect many aspects of the political and socioeconomic climate in which they operate Page 12 may partially explain the low levels of trust in the police expressed by citizens from post-socialist countries. DETERMINANTS OF PUBLIC ATTITUDES TOWARD THE POLICE In the past two decades, in particular, considerable research has been conducted to determine the basis for public attitudes toward the police and a diversity of concepts have been used to measure the residents’ attitudes toward police. For instance, researchers examined ‘sentiments toward police’, ‘confidence in the police’, ‘satisfaction with the police’, ‘support for the police’ or explored the public perceptions of ‘police performance’. Although most studies focused on general attitudes toward the police, there are also studies that examined specific public attitudes by assessing the quality of the respondent’s contact with the police. As a result of these methodological differences, research findings regarding public attitudes towards the police have been varied (Dowler & Sparks, 2008). Nevertheless, cross-national studies based on survey data usually considered that variations in certain sociodemographic indicators (eg, the respondent’s age, gender, race/ethnicity and social class), contextual variables (eg, the respondent’s place of residence and neighbourhood characteristics) and the quality of the citizens’ experience with police will generate variations in attitudes toward the police (see Dowler & Sparks). In terms of demographic characteristics, studies conducted in the USA found a significant relationship between age and attitudes toward the police. Specifically, younger people tended to have more negative attitudes toward the police than older residents (Dowler & Sparks, 2008; Reisig & Parks, 2000). Regarding the impact of gender, some studies found no significant relationship between gender and satisfaction Andreescu and Keeling with police (Frank et al., 2005), whereas other studies found that women held more positive attitudes toward the police than men did (Dowler & Sparks; Hawdon & Ryan, 2003). Although many studies conducted in the USA found that ethnic/racial minorities have a significantly lower level of satisfaction with the police than white residents do (Carter, 2002; Dowler & Sparks; Frank et al.; Reisig & Parks; Weitzer & Tuch, 2005), in an examination of rural juveniles’ attitudes toward the police, Hurst (2007) acknowledged that race was not a significant predictor of satisfaction with the police. Mixed results were also obtained when the effect of social class, usually operationalised in terms of income and education, has been considered. Frank et al. found that Americans belonging to the lowest income group had the lowest level of satisfaction with the police. Other studies, however, failed to find a significant relationship between income and attitudes toward the police (Ho & McKean, 2004). While Frank et al. found that more educated citizens express higher levels of satisfaction with the police Ho and McKean found no significant relationship between education and attitudes toward the police. Survey results indicate that ‘people who believe that in general most other people in their society can be trusted are also more inclined to have a positive view of their democratic institutions’ (Rothstein & Uslaner, 2005, p. 41). Hawdon and Ryan (2003) found that residents of neighbourhoods characterised by good interpersonal relationships, cohesion and community solidarity also expressed higher levels of satisfaction with police work. In the USA, researchers found that people who expressed fear of crime and a low sense of neighbourhood safety were less likely to have positive attitudes toward the police (Dowler & Sparks, 2008; Hawdon & Ryan, 2003; Tewksbury & West, 2001; Weitzer & Tuch, 2005). In general, findings suggest that residents of rural and smaller urban settings have a more positive attitude toward the police than those living in large urban areas (see Dowler & Sparks). Research in the USA has not substantiated a relationship between victimisation and attitudes toward the police (see Dowler & Sparks, 2008). Some studies (Hawdon & Ryan, 2003; Tewksbury & West, 2001) found that victims of crime had positive attitudes toward the police, whereas other studies (Dowler & Sparks) found a negative relationship between experience with victimisation and satisfaction with police work. Researchers also acknowledged that the level of public trust in the police was influenced by the quality of public–police encounters. An examination of data collected from residents of Houston, Texas showed a decrease in the level of confidence toward the police with an increase in the number of recent contacts (of any kind) a person had with the police. Nevertheless, negative experiences with the police reduced the public’s confidence in law enforcement more than positive contacts with the police did (Skogan, 2009). These findings are consistent with the results obtained in other research studies conducted in the USA and the UK, which concluded that the most positive attitudes toward the police were expressed by persons who had no recent experiences with them (Bradford et al., 2009; Skogan, 2006). In Finland, Kaariainen (2008) also found that trust in the police decreases with an increase in the number of negative experiences. However, the author observed that several negative encounters with the police did not lead to a significant reduction in one’s level of trust. In addition, Kaariainen acknowledged that, based on the Finnish data, trust does not seem to be associated at all with having had or not having had contact with the police. Page 13 Explaining the public distrust of police in the newest EU countries In a recent examination of correlates of trust in the police in 16 European countries, Kaariainen (2007) acknowledged that personal experiences of corruption among public officials and perceived corruption at the government level lessened the residents’ trust in the police. However, although it is generally hypothesised that direct experience with police corruption would negatively influence public perceptions of the police, Tankebe (2010), who recently conducted an empirical study in Accra, the capital of Ghana, found that, whereas vicarious experience of corruption and satisfaction with anticorruption policies explained variations in perceptions of police trustworthiness, personal experience of police corruption did not significantly affect one’s level of confidence in the police. Results showed that contained and low level corruption that followed wellestablished informal rules was tolerated by some segments of the population without diminishing one’s level of trust in the police (Tankebe). Regarding demographic determinants of attitudes toward the police in Europe, Kaariainen (2007) found that when controlling for a set of individual-level indicators, age was positively related to trust in the police, that European women tended to have a significantly higher level of trust in the police than men, and that residents of towns, smaller cities and villages expressed a higher level of trust in the police than those living in big cities. In addition, Kaariainen noted that Europeans experiencing serious financial difficulties tended to express a lower level of trust in the police. Comparable results were obtained by recent research conducted in the UK. Specifically, less favourable views of the police were expressed by young persons, whites, males, residents of inner-city areas and persons living on a lower income (Jackson et al., 2009). Page 14 Similar to findings in the USA, Europeans who lacked a sense of safety and felt insecure in their communities tended to have lower levels of confidence in the police (Jackson et al., 2009; Kaariainen, 2007). However, data from the UK showed that the overall confidence in local policing was influenced more by concerns about neighbourhood disorder than by fear of crime (Jackson et al.; Jackson & Sunshine, 2007). In their analysis of data from the British Crime Survey, Jackson et al. found that persons having a recent victimisation experience had less favourable views of the police. On average, in Europe, persons who had been victims of crime and those who belong to social minority groups trusted the police more than, respectively, those who did not experience victimisation or those who did not consider themselves members of a discriminated group (Kaariainen). Kaariainen also noted that variations in trust in the police are likely to be higher within a particular country than they are among countries because citizens’ trust in the police is a reflection of individual-level experiences, which can vary greatly. To summarise, prior research has demonstrated that both community-level and individual-level variables play a significant role in explaining variations in attitudes toward the police. Based on the reviewed literature, we expect that in Romania and Bulgaria, distrust of the police will correlate negatively with general institutional trust and interpersonal trust. We hypothesise that perceived economic exclusion and social discrimination will be positively related to distrust of the police. Similarly, we anticipate that persons who experienced victimisation (directly or indirectly) and those who expressed a high level of fear of crime will be less likely to trust the police. In addition, we explore variations in attitudes toward the police by gender, age, ethnic minority status, education, residency and region. Andreescu and Keeling DATA AND METHODS The source of the data was the European Social Survey (ESS), Round 4 (2008/2009). Data analysis was conducted on probability samples of persons aged 15 years and older from Romania (N = 2,101) and Bulgaria (N = 2230). The main purpose of this secondary analysis was to identify what characteristics have the persons who express distrust toward the police in Bulgaria and Romania. The variables used in the statistical analyses are briefly presented below. The dependent variable (police distrust) is a dummy variable created by dichotomising the original variable that measured the respondent’s level of trust in the police. The original variable had values from 0 (no trust at all) to 10 (complete trust). Respondents who expressed complete lack of trust and a low level of trust (ie, codes 0 to 3) were coded 1, whereas all the other responses were coded 0. Theoretical and methodological considerations were taken into account when the original variable has been transformed. Following Lucas (1980) and Mill (1948) who, inspired by Aristotle’s discussion of justice, acknowledged that many moral attributes are best defined by their opposites, we extrapolated and considered that ‘distrust’ instead of ‘trust’ should be the focus of this analysis. In addition, from a public policy point of view, it can be argued that it would be useful to identify the respondents who expressed distrust because they should be targeted if an improvement in public attitudes toward the police is to be obtained. Also, in both samples, the original variable did not have a normal distribution and was positively skewed, restricting the choice of statistical procedures that could be used to analyse the data. For instance, in the Bulgarian sample, the modal category was zero (ie, 23 per cent of the respondents expressed complete lack of trust toward the police), the median value was 3, and the mean was 3.21 (eg, 59 per cent of the respondents had scores of = 3). Although in the Romanian sample the mean value was higher (4.35), and a smaller proportion of respondents (40 per cent) had scores of = 3, the same transformation was used to permit valid intersample comparisons. It should be noted that in both samples distrust (codes 0–3) was more prevalent than scepticism (codes 4–6) or trust toward the police (codes 7–10). A similar three-way distinction in levels of trust toward the police was made by Mishler and Rose (1997). Institutional trust is a composite indicator that includes five highly correlated variables that measure the respondents’ level of satisfaction with the national governments and their trust in the legal system, the country’s parliament, political parties and politicians. All forming indicators were continuous variables that took values from 0 (lowest level of trust/satisfaction) to 10 (highest level of trust/satisfaction). A factor measuring institutional trust was created based on principal component analysis (PCA). The index appears to be a reliable and valid measure of institutional trust (ie, Romanian sample: Cronbach’s coefficient alpha = 0.912; Bulgarian sample: Cronbach’s coefficient alpha = 0.911). In PCA only one component was extracted (Romanian sample: Eigenvalue = 3.722; variance explained = 74.43 per cent; Bulgarian sample: Eigenvalue = 3.729; variance explained = 74.59 per cent) and factor loadings varied from 0.718 to 0.928 (Romanian sample) and 0.775 to 0.898 (Bulgarian sample). These results indicate that the index of institutional trust is a unidimensional measure (eg, persons with high level of trust in one institution tend to have high levels of trust in the other institutions listed here). Higher factor scores indicate a higher level of institutional trust. Interpersonal trust is also a composite measure and includes three highly correlated indicators. The index combines responses at three questions that asked re- Page 15 Explaining the public distrust of police in the newest EU countries spondents to express opinions about their fellow citizens (ie, how trustworthy, helpful and fair people are) and rank them on a scale from 0 (negative opinion) to 10 (positive opinion). Factor analysis (PCA) was used to compute the index. When PCA was conducted only one factor was extracted in each sample (Romanian sample: Eigenvalue = 2.310; variance explained by the factor = 77 per cent; Bulgarian sample: Eigenvalue = 2.182; variance explained by the factor = 73 per cent) and factor loadings for individual variables ranged from 0.856 to 0.901 (Romanian sample) and 0.826 to 0.879 (Bulgarian sample). The Cronbach’s alpha coefficients of reliability for this indicator were 0.854 (Romanian sample) and 0.812 (Bulgarian sample). Relatively large alpha values and the results of the factor analysis indicate that the composite measure is a unidimensional index of interpersonal trust. Higher factor scores indicate a higher level of interpersonal trust. Experience with victimisation is a binary variable, coded 1 for respondents who have been, themselves or a member of their family, victim(s) of a violent crime (robbery or assault) in the past five years and 0 otherwise. In addition, we used a fear of crime index as a predictor of trust in the police. The composite indicator was calculated based on factor analysis and incorporates three indicators of fear of crime. The questions used in this analysis are: (1) How safe do you feel walking alone in local area after dark? (2) How often do you worry about your home being burgled? (3) How often do you worry about becoming a victim of a violent crime? When PCA was conducted, only one factor was extracted in each sample (Romanian sample: Eigenvalue = 1.999; variance explained by the factor = 66.6 per cent; Bulgarian sample: Eigenvalue = 2.182; variance explained by the factor = 72.7 per cent), suggesting that persons who are afraid of walking alone at night are also more likely to be afraid of Page 16 becoming a victim of a violent or property crime. Factor loadings varied from 0.727 to 0.859 (Romanian sample) and 0.806 to 0.878 (Bulgarian sample). The reliability coefficients alpha were 0.747 (Romanian sample) and 0.810 (Bulgarian sample). Higher factor scores indicate a higher level of fear of crime. Relatively large alpha values and the results of the factor analysis indicate that the computed index is a unidimensional measure of fear of crime. Research has shown that individuals who perceive themselves as being socially and/or economically excluded from a society (see Frank et al., 2005; Kaariainen, 2007) will express a lower level of trust in official authorities, police included. The measure of perceived social exclusion used in this analysis is a dummy variable (ie, respondents who declared they belong to a discriminated group were coded 1, the others were coded 0). The social exclusion indicator referred to perceived discrimination based on ethnicity, age, gender, sexual orientation, language or religion. In addition, we computed a dummy variable for perceived economic exclusion. Respondents who expressed financial difficulties and declared they find it ‘very difficult to live on the household income’ were coded 1, the others were coded 0. Although perceptions of corruption and experience with corruption are considered to be important predictors of institutional trust (Kaariainen; Tankebe, 2010; Uslaner, 2010), we could not determine whether in Romania and Bulgaria attitudes toward the police are influenced by public perceptions of police misconduct or experiences with corrupt behaviour because ESS4 did not include questions that would measure these indicators. As control variables we used age, gender, education, ethnic minority status, region of residency and residential area (urban vs. rural). Age is a dummy variable, coded 1 for respondents belonging to the age group Andreescu and Keeling 55–64. Additional analyses (not included) showed that people in this particular age group expressed the lowest level of trust in the police. Gender is a dummy variable coded 1 for females and 0 for males. Education is a continuous variable that measures the number of years of school completed. Ethnic minority — respondents who stated that they belong to an ethnic group were coded 1, the others were coded 0. Residential area — respondents who were residents of rural areas (villages, farms) were coded 1, the others were coded 0. Preliminary analyses showed that in Romania, among eight identified regions included in the data set, residents from Bucharest (capital city of Romania) expressed the lowest level of trust in the police. Similar attitudes were expressed by residents of Sofia, the capital city of Bulgaria. The variable region is a dummy variable coded 1 for residency in the capital city and 0 otherwise. RESULTS In 2007, based on data from the European Sourcebook of Crime and Criminal Justice Statistics (Killias et al., 2010), crime rates for a selected group of offences (ie, murder, assault, rape, robbery, theft, motor vehicle theft and drug offences) were lower in Bulgaria and Romania than the corresponding figures for Europe (see Table 1). Data suggest not only that the level of criminality was, on average, lower in these two countries than in Europe as a whole, but also that the local police seemed to be relatively more effective in preventing criminal behaviour and securing public safety. However, the actual crime rates do not appear to influence public attitudes toward the police in Romania and Bulgaria. Preliminary analyses of ESS4 (ESS, 2008) country survey data referring to residents’ trust in police showed that in Romania and Bulgaria the levels of public confidence in local police were among the lowest in the EU. In 2008, the average level of public trust in the police, measured on a scale from 0 (complete distrust) to 10 (complete trust), varied from 3.21 in Bulgaria to 7.99 in Finland. In Romania, the calculated mean value (4.35) was the second lowest among EU states. Approximately 40 per cent of Romanians and 59 per cent of Bulgarians expressed a low level of trust in the police. Nevertheless, in both countries, the public trust in police appears to be higher than the residents’ level of confidence in other important societal institutions, such as the Table 1: Crimes recorded by the police in Romania and Bulgaria in 2007 Crime rate per 100,000 people Murder Assault Rape Robbery Theft Motor vehicle theft Drug offences Total crimes Romania Bulgaria Europe (Average) 3.9 43 5 11 199 8 13 1263.5 3.4 43 3 41 883 6 39 1839.2 4.9 229 9 56 1770 131 203 NA Source: Killias et al. (2010). Page 17 Explaining the public distrust of police in the newest EU countries Table 2: Institutional trust in Romania and Bulgaria Romania (N = 2,101) Trust in the police Trust in country’s Parliament Trust in the legal system Satisfaction with the government Trust in political parties Trust in politicians Bulgaria (N = 2,230) Mean % (Very) low Mean % (Very) low 4.35 3.82 3.78 3.52 3.13 3.05 40.3 47.3 47.2 50.0 58.0 60.7 58.8 78.4 74.4 75.3 81.9 82.8 3.21 1.88 2.23 2.23 1.70 1.61 Note: (Very) low level of trust refers to the percentage of respondents who selected responses coded 0–3. legal system, the country’s parliament, the national government, the main political parties or politicians (Table 2). Data included in Table 2 suggest that the lack of institutional trust is more pronounced in Bulgaria than in Romania. The present quantitative analysis attempts to determine if the respondents’ level of trust in legal and political institutions, interpersonal trust, fear of crime, perceived social and economic exclusion, and experience with victimisation significantly influence public attitudes toward the police. In addition, age, gender, education and residency will be used as control variables. The descriptive statistics for all the variables included in multivariate analyses are presented in Table 3. Binomial logistic regression was used to identify the variables able to predict distrust Table 3: Descriptive statistics Romania (N = 2,101) Distrust in the police Institutional trust Interpersonal trust Victimisation Fear of crime Social exclusion Economic exclusion Age (55–65 years) Gender (female) Education Ethnic minority Residency (rural) Capital city resident Mean SD 0.40 0.00 0.00 0.10 0.00 0.04 0.20 0.16 0.54 110.3 0.16 0.46 0.11 0.49 10.00 10.00 0.30 10.00 0.20 0.40 0.37 0.59 30.81 0.36 0.49 0.31 Note: Reliability coefficients alpha reported for the composite measures. Page 18 Bulgaria (N = 2,230) Alpha 0.912 0.850 0.747 Mean SD 0.59 0.00 0.00 0.15 0.00 0.10 0.35 0.20 0.56 110.0 0.18 0.34 0.16 0.49 10.00 10.00 0.36 10.00 0.30 0.48 0.41 0.50 30.64 0.38 0.47 0.37 Alpha 0.911 0.812 0.810 Andreescu and Keeling factor of 0.211 [Exp(B) = 0.211] in Bulgaria. These results suggest that Romanians and Bulgarians who trust the legal system, the country’s parliament, the political parties, politicians in general and those who are satisfied with the national government are more likely to express trust in the police as well. Beside institutional trust, age is the second variable that appears to influence attitudes toward the police in Romania, while fear of crime and ethnic minority status are significantly predicting variations in trust toward the police in Bulgaria. Exploratory analyses did not indicate a linear relationship between age and trust in the police. Although bivariate analyses showed that in both countries persons age 55–64 had the lowest average level of confidence in the police, age had a significant effect in multivariate analyses only in Romania. Model 1 predicts that, compared with people in other age groups, the odds for Romanian in the police in Romania and Bulgaria. Table 4 presents the results for the first competing model that includes all the variables in the equation. It should be noted that the logistic regression Model 1 fits the data well. Based on Hosmer and Lemeshow goodness-of-fit tests (eg, for the Romanian sample χ2 = 10.97, df = 8, p = 0.20; for the Bulgarian sample χ2 = 5.164, df = 8, p = 0.74), we cannot reject the null hypothesis that there is no difference between the observed and predicted values of the dependent variable. The results imply that the model’s estimates fit the data at an acceptable level for both samples. When controlling for a set of indicators (Table 4, Model 1), institutional trust is significantly and negatively related to the dependent variable. Specifically, if institutional trust increases one unit, the odds of distrusting the police decrease by a factor of 0.150 [Exp(B) = 0.150] in Romania and by a Table 4: Logit estimates for police distrust in Romania and Bulgaria (Model 1) Romania (N = 2,101) B Institutional trust Interpersonal trust Victimisation Fear of crime Economic exclusion Social exclusion Ethnic minority Gender (F) Age (55–64 years) Education Residency (rural) Capital city (Bucharest/Sofia) Constant Nagelkerke R2 Goodness-of-fit –1.898 0.128 0.001 0.055 0.097 –0.191 –0.102 0.064 0.601 –0.029 0.035 0.186 –0.447 0.478 Bulgaria (N = 2,230) p Exp(B) B SE p Exp(B) 0.103 0.074 0.220 0.074 0.177 0.339 0.198 0.137 0.184 0.020 0.147 0.218 0.299 0.000 0.084 0.998 0.456 0.582 0.574 0.607 0.642 0.001 0.151 0.811 0.392 0.134 0.150 1.137 1.001 1.057 1.102 0.826 0.903 1.066 1.824 0.971 1.036 1.205 0.640 –1.554 –0.076 –0.149 0.166 0.029 0.397 –0.374 0.133 0.161 0.013 0.189 0.235 0.165 0.407 0.083 0.064 0.165 0.068 0.138 0.222 0.187 0.123 0.146 0.021 0.142 0.172 0.293 0.000 0.239 0.368 0.014 0.834 0.074 0.045 0.280 0.269 0.521 0.184 0.172 0.574 0.211 0.927 0.862 1.181 1.029 1.487 0.688 1.142 1.175 1.014 1.208 1.265 1.179 χ2 10.971 df 8 p 0.20 χ2 5.164 df 8 p 0.74 SE Page 19 Explaining the public distrust of police in the newest EU countries respondents age 55–64 to express distrust in the police are 82 per cent higher [Exp(B) = 1.824], when controlling for the other variables in the model. Although fear of crime is positively related to police distrust in both countries, only in Bulgaria is the effect significant, suggesting that for each unit increase in one’s level of fear of crime, the odds of distrusting the police increase by 18 per cent. Although ethnic minorities in both countries are less likely to distrust the police, the effect is significant only in Bulgaria. A second competing logistic regression model is presented in Table 5. This model does not include the variable institutional trust. It can be observed that without institutional trust in the model, the strength of the association between the predictors and the dependent variable is substantially reduced (ie, Nagelkerke R2 decreased from 0.478 to 0.063 in the Romanian sample and from 0.407 to 0.109 in the Bulgarian sample), suggesting that trust in other important institutions is the most influential predictor of attitudes toward the police in Romania and Bulgaria. Goodness-of-fit tests indicate that Model 2 fits the data adequately for both samples. There are no significant differences between the observed values and the values of the dependent variable predicted by this model (ie, Romanian sample χ2 = 8.990, df = 8, p = 0.38; Bulgarian sample χ2 = 14.022, df = 8, p = 0.08). As hypothesised, in both countries, with an increase in the respondents’ level of interpersonal trust there is a significant decrease in the respondents’ distrust of the police. For instance, results show that the odds of distrusting the police are approximately 20 per cent lower (ie, 1 – 0.798 = 0.202) in Romania and 31 per cent lower in Bulgaria (ie, 1 – 0.688 = 0.312) for persons whose level of interpersonal trust is one standard deviation above the group mean Table 5: Logit estimates for police distrust in Romania and Bulgaria (Model 2) Romania (N = 2,101) Institutional trust Interpersonal trust Victimisation Fear of crime Economic exclusion Social exclusion Ethnic minority Gender (F) Age (55–64 years) Education Residency (rural) Capital city (Bucharest/Sofia) Constant Nagelkerke R2 Goodness-of-fit Page 20 Bulgaria (N = 2,230) B SE p Exp(B) B SE p Exp(B) — –0.226 0.084 0.176 0.260 –0.109 –0.188 –0.134 0.420 0.002 –0.175 0.502 –0.416 0.063 — 0.053 0.169 0.056 0.132 0.265 0.154 0.105 0.139 0.015 0.112 0.172 0.227 — 0.000 0.619 0.002 0.050 0.681 0.221 0.202 0.003 0.909 0.119 0.004 0.067 — 0.798 10.088 10.192 10.297 0.897 0.828 0.874 10.522 10.002 0.839 10.652 0.660 — –0.375 0.016 0.205 0.232 0.511 –0.901 –0.196 0.131 0.018 –0.074 0.199 0.234 0.109 — 0.049 0.135 0.054 0.110 0.176 0.144 0.099 0.117 0.016 0.113 0.148 0.231 — 0.000 0.907 0.000 0.034 0.004 0.000 0.047 0.263 0.271 0.511 0.150 0.311 — 0.688 1.016 1.227 1.262 1.667 0.406 0.822 1.140 1.018 0.929 1.220 1.263 χ2 8.990 df 8 p 0.38 χ2 14.022 df 8 p 0.08 Andreescu and Keeling compared with those whose level of interpersonal trust is equal to the group mean. The variables fear of crime and economic exclusion have also significant effects in both countries. Although the results indicate that prior direct or indirect experience with victimisation did not have a significant impact on trust in the police in Romania and Bulgaria, consistent with our predictions, persons who feel less safe in their neighbourhoods, who think frequently that they might be burgled or become a victim of a violent crime are more likely to distrust the police. For one unit increase in one’s level of fear of crime, the odds of distrusting the police increase by a factor of 1.19 in Romania and by a factor of 1.23 in Bulgaria. When controlling for the other variables in the model, the odds of one distrusting the police are in Romania almost 30 per cent [Exp(B) = 1.297] and in Bulgaria 26 per cent [Exp(B) = 1.262] higher for people with insufficient household incomes compared with persons who are better off economically. Approximately 35 per cent of the Bulgarians and 20 per cent of Romanians expressed financial difficulties (see Table 3) in 2008. Even if the direction of the effects is almost identical when samples are compared, intercountry differences in the magnitude of the effects can be observed. Whereas age and residency in capital city are significantly associated with police distrust in Romania, in Bulgaria, perceived social exclusion is more likely to be associated with a lack of confidence in the police. Specifically, the odds of one distrusting the police are 52 per cent higher for a Romanian aged 55–64 and 65 per cent higher for a Bucharest resident when compared with persons belonging to other age groups or those who are not capital-city residents, respectively. In Bulgaria, women and members of ethnic minority groups are less likely to distrust the police when compared with men and members of the majority population, respectively. However, Bulgarians who perceive themselves as being discriminated against based on ethnicity, age, gender, sexual orientation, language or religion are 67 per cent more likely to express distrust of the police than persons who do not feel socially excluded. Perceived discrimination is acknowledged by 10 per cent of the Bulgarians and only by 4 per cent of the Romanians (see Table 3). Even if in both countries the lack of confidence in the police tends to increase with educational level, the effects were not sufficiently strong to be significant. Similarly, although residents of rural areas in both countries were more likely to trust the police, their opinions about the police were not significantly different from those expressed by residents of urban or suburban areas. DISCUSSION AND CONCLUSIONS Consistent with prior research (see Kaariainen, 2007; Zvekic, 1996), the present analysis showed that the average postaccession level of public trust in the police in the newest EU member states continued to be lower than in Western European democracies and also lower than in other former communist countries that joined the EU in 2004. Similar to findings from other studies (see Kaariainen; Rothstein, 2005) and as hypothesised, institutional and interpersonal trust were significantly related to public trust in the police in Romania and Bulgaria. In both countries, the variable found as having the greatest influence on a person’s level of trust in police is institutional trust. Even if it can be considered that a pervasive culture of distrust and a higher level of social pessimism partially influenced Bulgarians’ and Romanians’ persistent low Page 21 Explaining the public distrust of police in the newest EU countries levels of confidence in governance and major institutions, findings based on recent national public opinion polls conducted in Romania and Bulgaria indicated that public discontent with national institutions, the courts and police included, was also influenced by the fact that major societal institutions were described by citizens as lacking transparency and accountability and were perceived as being highly corrupt (Center for the Study of Democracy, 2011; Deleanu & Bechir, 2011; Slavova, 2010). As found in prior research (Dowler & Sparks, 2008; Kaariainen, 2007; Reisig & Parks, 2000; Weitzer & Tuch, 2005), negative public attitudes toward the police in Bulgaria and Romania appear to be also related to residents’ fears of victimisation. Although law enforcement officers can do little to change subjective opinions about perceived safety, research has shown that negative personal experiences with law enforcement had detrimental effects on the level of public confidence in the police (see Reisig & Parks; Skogan, 2006, 2009). In addition, ‘by actively engaging a public relations strategy to fight crime or reduce crime rates’, the police may be able to improve the citizens’ perceptions of the crime problem in their neighbourhood, which could eventually have a positive impact on people’s satisfaction with local police (Dowler & Sparks). Research also suggests that public confidence in policing is less about the effectiveness of police services and more about procedural fairness of the police in contacts with the public and the police capacity to identify with the values of the group they are supposed to protect (see Jackson et al., 2009). Unlike in other studies, direct or indirect experiences with victimisation do not appear to have a significant influence on public attitudes toward the Romanian or Bulgarian police. This finding could be a result of the respondents’ different experiences with police handling their cases, if Page 22 they were victimised. For instance, victims whose cases had been cleared by the police probably had reasons to trust the police, whereas those whose cases remained unsolved probably had the opposite attitude toward the police, cancelling the effect of positive attitudes expressed by others. Also different from research conducted in other countries, particularly the USA (Carter, 2002; Dowler & Sparks, 2008; Frank et al., 2005; Reisig & Parks, 2000; Weitzer & Tuch, 2005), ethnic minorities, especially in Bulgaria, appear to have a higher level of confidence in the police than the majority population. These results are, however, consistent with Kaariainen’s (2007) findings in Europe as a whole and with data from the UK, where recent opinions of the police appear to be higher among most minority groups than they are in the majority population (Bradford & Jackson, 2011). Similar to other studies conducted in Europe (Jackson et al., 2009; Kaariainen), Bulgarian and Romanian women tend to have higher levels of confidence in the police than their male counterparts. Although this study does not identify a linear relationship between age and attitudes toward the police, as many prior studies did, results show that, particularly in Romania, individuals who are approaching retirement age are the most distrustful of the police. It is possible that persons aged 55–64, who were relatively young when the regime changed more than two decades ago, had higher expectations regarding the new democratic institutions that did not materialise, contributing now to their higher level of social scepticism toward governance and lower levels of institutional trust. As anticipated and consistent with past research (Frank et al., 2005; Jackson et al., 2009; Kaariainen, 2007), in both countries, people who perceive themselves as economically marginalised are less likely to Andreescu and Keeling trust the police and other societal institutions. Similar attitudes are expressed by Bulgarian residents who consider themselves socially excluded. As Goldsmith (2005) observed, persons who are socially disadvantaged often have distinct life experiences that negatively affect their perceptions of public services, police included. Although this study provides information about the relative influence of certain factors that account for variation in the public’s opinion about law enforcement in Romania and Bulgaria, this research is a secondary analysis limited by the existent data, which did not include potentially important indicators. Although the findings suggest that the level of trust in the police mirrors at a certain degree the generalised institutional trust expressed by nationals of Bulgaria and Romania, future research should include additional factors to better explain variations in public attitudes toward the police. For instance, based on the available data, we had no way of knowing why respondents had a certain degree of confidence in the police. In addition to larger social forces, public attitudes toward the police may be influenced by the police conduct as well. Analyses of crime victimisation surveys showed that at the world level there is a strong positive correlation between public satisfaction with the police and the frequency of police patrolling, the police ability to prevent and control crime, and their efficiency in handling crime reports (Zvekic, 1996). However, ESS4 did not include information regarding police performance, police misconduct or public experiences with corruption and the effect of these indicators on public trust in the police could not be examined. Without a doubt, there are inherent constraints upon trust-building in postcommunist Romanian and Bulgarian societies and the impact of the wider political and social environment in which policing occurs cannot be neglected. Nevertheless, the restoration of public trust in police is attainable and should be regarded as an important part of democratic policing in these countries. Sztompka (1998) argues that the emergence of a culture of trust is possible in post-socialist societies because the new democratic organisations have the capacity to create trust-generating conditions, such as normative certainty, transparency, stability and accountability. Phillips and Trone (2002) suggest that by exposing police practices, pointing out the shortcomings in how police regulate themselves, reporting honestly on the depth and pace of police reform, and engaging the public and the police in a dialogue, the citizens’ perception of the institution could improve. O’Neill (2002), however, emphasises the fact that if, in theory, the new culture of accountability and audit makes professionals and institutions more accountable to the public, it also provides incentives for arbitrary and unprofessional choices. For instance, police procedures for preparing cases are so demanding now that fewer cases can be prepared and fewer criminals are brought to court. The author argues that intelligent accountability that requires more attention to good governance should replace excessive regulation, micromanagement and central control. Similarly, O’Neill acknowledges that demands for universal transparency are likely to encourage evasions, hypocrisies and half-truths. In the O’Neill’s view, global transparency and complete openness are not the best ways to build or restore trust; the focus should be on limiting deception. In a culture of public service, ‘professionals and public servants should be free to serve the public rather than their paymasters’. REFERENCES Abraham, P. (2001). The Romanian police and challenges of transition. In A. Kadar (Ed.), Page 23 Explaining the public distrust of police in the newest EU countries Police in transition. Essays on the police forces in transition countries (pp. 115–129). Budapest, Hungary: Central European University Press Ackerman, G., Anderson, B., Jensen, S., Ludwig, R., Montero, D., Plante, N., & Yanez, V. (2001). Crime rates and confidence in the police: America’s changing attitudes toward crime and police, 1972–1999. Journal of Sociology and Social Welfare, 28, 43–54. Aromaa, K., Leppa, S., Nevala, S., & Ollus, N. (Eds.). (2003). Crime and criminal justice in Europe and North America, 1995–1997. Helsinki, Finland: UN European Institute for Crime Prevention and Control. Barbalet, J. (2009). A characterization of trust, and its consequences. Theory and Society, 38, 367–382. Benson, P. R. (1981). Political alienation and public satisfaction with police services. Pacific Sociological Review, 24, 45–64. Bojkov, V. (2006). Integrating the Bulgarian security sector in Euro-Atlantic structures: post1989 reforms in a changing landscape of (in)securitisation. LSE Challenge Papers, Work Package 2: Securitization beyond borders. London: London School of Economics. Retrieved January 11, 2012, from http:// www2.lse.ac.uk/internationalRelations/ centresandunits/EFPU/EFPUchallenge papers.aspx Bradford, B., & Jackson, J. (2011). Why Britons trust their police. Books and ideas. Retrieved January 10, 2012, from http:// www.booksandideas.net/IMG/pdf/ Bradford_Jackson_EN.pdf Bradford, B., Jackson, J., & Stanko, E. A. (2009). Contact and confidence: revisiting the impact of public encounters with the police. Policing & Society, 19, 20–46. Caparini, M. & Marenin, O. (2005). Crime, insecurity and police reform in post-socialist CEE. The Journal of Power Institutions in PostSoviet Societies, 2. Retrieved August 11, 2011, from http://pipss.revues.org/330. Carter, D. L. (2002). The police and the community (7th ed.). Englewoods Cliffs, NJ: Prentice Hall. Page 24 Center for the Study of Democracy (2011). Public trust in the criminal justice system – an instrument for penal policy assessment, Policy Brief No. 29. Sofia, Bulgaria: CSD. Retrieved July 1, 2011, from http://www. csd.bg/artShow.php?id=15598 Deleanu, V., & Bechir, M. (2011, February 3). Sondaj IMAS: Românii au tot mai put,inã încredere în Polit,ie. Adevarul. Retrieved May 7, 2011, from http://www.adevarul.ro /actualitate/eveniment/Politia-amendata_ de_popor_0_436157052.html. Devos, T., Spin, D., & Schwartz, S. H. (2002). Conflicts between human values and trust in institutions. British Journal of Psychology, 41, 481–484. Dowler, K., & Sparks, R. (2008). Victimization, contact with police, and neighborhood conditions: reconsidering African Americans and Hispanic attitudes toward the police. Police Practice & Research: An International Journal, 9, 395–415. European Commission. (2011, July 20). Report from the Commission to the European Parliament and the Council: on progress in Romania under the cooperation and verification mechanism. Retrieved January 10, 2012, from http:// ec.europa.eu/dgs/secretariat_general/cvm/ docs/com_2011_460_en.pdf European Commission (2012a, February 8). Report from the Commission to the European Parliament and the Council: interim report on progress in Romania under the co-operation and verification mechanism. Retrieved February 12, 2012, from http://www.just.ro/Portals/0/Comun icate/Februarie per cent202012/com_ 2012_56_en.pdf European Commission (2012b, February 8). Report from the Commission to the European Parliament and the Council: interim report on progress in Bulgaria under the co-operation and verification mechanism. Retrieved February 12, 2012, from http://ec.europa.eu/cvm/docs/com_ 2012_57_en.pdf European Social Survey (ESS). (2008). Round 4 data. Data file edition 1.0. Bergen, Andreescu and Keeling Norway: Norwegian Social Science Data Services (data archive and distributor of ESS data). Frank, J., Smith, B. W., & Novak, K. J. (2005). Exploring the basis of citizens’ attitudes toward the police. Police Quarterly, 8, 206–228. Goldsmith, A. (2005). Police reform and the problem of trust. Theoretical Criminology, 9, 443–470. Guvernul Romaniei. (2005, September 7). Codul de etica si deontologie al politistului. Decizia 991, 25.08.2005. Monitorul Oficial, No. 813. Retrieved January 10, 2011, from http://www.mai.gov.ro/Documente/Cariera /Cod_etica.pdf Hawdon, J., & Ryan, J. (2003). Police–resident interactions and satisfaction with police: an empirical test of community policing assertions. Criminal Justice Policy Review, 14, 55–74. Hinton, M. S., & Newborn, T. (2009). Introduction. In M. S. Hinton & T. Newborn (Eds.), Policing developing democracies (pp. 1–28). New York, NY: Routledge. Ho, T., & McKean, J. (2004). Confidence in the police and perceptions of risk. Western Criminological Review, 5, 108–118. Hough, M., Jackson, J., Bradford, B., Myhill, A., & Quinton, P. (2010). Procedural justice, trust, and legitimacy. Policing: A Journal of Policy and Practice, 4, 203–210. Hurst, V. G. (2007). Juvenile attitudes toward the police. Criminal Justice Review, 32, 121–141. Jackson, J., & Bradford, B. (2009). Crime, policing and social order: on the expressive nature of public confidence in policing. British Journal of Sociology, 60, 493–521. Jackson, J., Bradford, B., Hohl, K., & Farrall, S. (2009). Does the fear of crime erode public confidence in policing? Policing: A Journal of Policy and Practice, 3, 100–111. Jackson, J., & Sunshine, J. (2007). Public confidence in policing: a neo-Durkheimian perspective. British Journal of Criminology, 47, 214–233. Kaariainen, J. T. (2007). Trust in the police in 16 European countries: a multilevel analysis. European Journal of Criminology, 4, 409–435. Kaariainen, J. T. (2008). Why do the Finns trust the police? Journal of Scandinavian Studies in Criminology and Crime Prevention, 9, 141–159. Kadar, A. (Ed.). (2001). Police in transition. Essays on the police forces in transition countries. Budapest, Hungary: Central European University Press. Killias, M., et al. (2010) European sourcebook of crime and criminal justice statistics – 2010. 285 Onderzoek en beleid (4th ed.). Den Haag, The Netherlands: Boom Juridische uitgevers. Ministry of Justice, Research and Documentation Centre (WODC). Retrieved March 2, 2011, from http:// europeansourcebook.org/ob285_full.pdf Koci, A. (1996). Legitimation and culturalism: towards policing changes in the European ‘post- socialist’ countries. In M. Pagon (Ed.), Policing in Central and Eastern Europe: comparing firsthand knowledge with experinece from the West. Ljubljana, Slovenia: College of Police and Security Studies. Retrieved July 5, 2012, from https://www.ncjrs.gov/ policing/leg219.htm Koszeg, F. (2001). Introduction. In A. Kadar (ed.) Police in transition. Essays on the police forces in transition countries (pp. 1–12). Budapest, Hungary: Central European University Press. Lucas, J. R. (1980). On justice. Oxford: Clarendon Press. Marinova, G. (2006). Bulgarian criminal procedure: the new philosophy and issues of approximation. Review of Central and East European Law, 31, 45–79. Maxson, C., Hennigan, K., & Sloane, D. C. (2003). Factors that influence public opinion of the police (NIJ Research for Practice, NCJ 197925). Washington, DC: US Department of Justice, National Institute of Justice. Mill, J. S. (1948). Utilitarianism, liberty and representative government. London: Dent, Everyman’s edition. Ministerul Administratiei si Internelor (MAI). (2011). Evaluarea Activitatii Desfasurate de Page 25 Explaining the public distrust of police in the newest EU countries Ministerul Administratiei si Internelor in Anul 2010. Bucuresti, Romania: Editura MAI. Ministry of Administration and Interior (MAI). (n.d.). Brief history. Retrieved September 1, 2011, from http://www.mai.gov.ro/engleza/ Home_eng/english.htm Ministry of the Interior (MoI). (n.d.). About the Ministry of the Interior. Retrieved January 10, 2012, from www.mvr.bg/en/ Mishler, W., & Rose, R. (1997). Trust, distrust and skepticism: popular evaluations of civil and political institutions in post-communist societies. Journal of Politics, 59, 418–451. Mobekk, E. (2005). Police reform in South East Europe: an analysis of the stability pact self-assessed studies. In E. Cole, T. Donais, & P. H. Fluri (Eds.), Defence and security sector governance and reform in South East Europe self-assessment studies: regional perspectives (pp. 155–167). Baden-Baden, Germany: Nomos Verlagsgesellschaft. Mueller, G. P. (2008). Trust and life satisfaction in Eastern and Western Europe. In V. Møller and D. Huschka (Eds.), Quality of life and the millennium challenge (pp. 161–176). Dordrecht, The Netherlands: Springer Myhill, A., & Quinton, P. (2010). Confidence, neighbourhood policing, and contact: drawing together the evidence. Policing: A Journal of Policy and Practice, 4, 273–281. Novinite. (2012, January 31). Bulgaria’s Interior Ministry to restrict use of force, firearms. Retrieved February 9, 2012. http://www.novinite.com/view_news. php?id=136240 O’Neill, O. (2002). A question of trust: the BBC Reith lectures 2002. Cambridge, UK: Cambridge University Press. Organization for Security and Co-operation in Europe (OSCE). (2009). POLIS – Policing profiles of participating and partner states: Bulgaria. Retrieved December 28, 2011, from http://polis.osce.org/countries/ details?item_id=11. Organization for Security and Co-operation in Europe (OSCE). (2010). POLIS – Policing profiles of participating and partner states: Romania. Retrieved December 28, 2011. from http://polis.osce.org/countries/ details?item_id=40. Page 26 Phillips, E., & Trone, J. (2002). Building public confidence in police through civilian oversight. New York, NY: Vera Institute of Justice. Politia Romana. (n.d.). Informatii Generale: Scurt istoric. Retrieved January 10, 2011, from http://politiaromana.ro/scurt_ istoric.htm Racovita, M. (2011). Europenization and effective democracy in Romania and Bulgaria. Romanian Journal of Political Science, 11, pp. 28–49. Reisig, M. D., & Parks, R. B. (2000). Experience, quality of life, and neighborhood context: A hierarchical analysis of satisfaction with police. Justice Quarterly, 17, 607–629. Reynolds, K. M., Semukhina, O. B., & Demidov, N. N. (2008). A longitudinal analysis of public satisfaction with the police in the Volvograd region of Russia. International Criminal Justice Review, 18, 158–189. Rothstein, B. (2005). Social traps and the problem of trust. Cambridge, UK: Cambridge University Press. Rothstein, B. & Stolle, D. (2003). Social capital, impartiality, and the welfare state: an institutional approach. In M. Hooghe & D. Stolle (Eds.), Generating social capital: the role of voluntary associations, institutions, and government policy (pp. 191–210). New York, NY: Palgrave Macmillan. Rzeplinski, A. (2001). The police in the constitutional framework: the limits of policing. In A. Kadar (Ed.), Police in transition. Essays on the police forces in transition countries (pp. 71–88). Budapest, Hungary: Central European University Press. Scaglion, R., & Condon, R. G. (1980). Determinants of attitudes toward city police. Criminology, 17, 485–494. Skogan, W. G. (2006). Asymmetry in the impact of encounters with police. Policing & Society, 16, 99–126. Skogan, W. G. (2009). Concern about crime and confidence in the police. Police Quarterly, 12, 301–318. Slavova, Z. (2010, January). Effectiveness of police work: strange disproportions. Politiki. Andreescu and Keeling Retrieved August 5 2011, from http:// politiki.bg/?cy=165&lang=2&a0i=223484 &a0m=readInternal&a0p_id=631. Spirova, M. (2009). Bulgaria. European Journal of Political Research, 48, 913–916. Sztompka, P. (1998). Trust, distrust and two paradoxes of democracy. European Journal of Social Theory, 1, 19–32. Sztompka, P. (1999). Trust: a sociological theory. Cambridge, UK: Cambridge University Press. Tankebe, J. (2010). Public confidence in the police: testing the effects of experience of police corruption in Ghana. British Journal of Criminology, 50, 296–319. Tewksbury, R., & West, A. (2001). Crime victims’ satisfaction with police services: an assessment in one urban community. The Justice Professional, 14, 271–285. Transparency International. (2010). Global corruption barometer 2010. Retreived September 9, 2011, from http://www. transparency.org/policy_research/surveys_ indices/gcb/2010 Tsvetkova, R. (2011, August 15). Priorities in the work of the Interior Ministry. Radio Bulgaria. Retrieved December 20, 2011, from http://bnr.bg/sites/en/Lifestyle/ Politics/Pages/1508Prioritiesintheworkof theInteriorMinistry.aspx. Uildriks, N., & van Reenen, P. (2003). Policing post-communist societies. Police–public violence, democraticpolicing and human rights. Antwerp, Belgium: Intersentia. Uslaner, E. M. (2002). The Moral Foundations of Trust. New York, NY: Cambridge University Press. Uslaner, E. M. (2005). Varieties of trust. FEEM Working Paper No. 69.05. Retreived January 10, 2012, from http://papers.ssrn. com/sol3/papers.cfm?abstract_id=726182. Uslaner, E. M. (2010). Corruption, inequality, and the rule of law. New York, NY: Cambridge University Press. Watts, L. L. (2004). Civil–military relationships. In H. F. Carey (Ed.), Romania since 1989: politics, economics, and society (pp. 523–551). Oxford, UK: Lexington Books. Weber, R. (2001). Police organization and accountability: a comparative study. In A. Kadar (Ed.), Police in transition. Essays on the police forces in transition countries (pp. 39– 69). Budapest, Hungary: Central European University Press. Weitzer, R., & Tuch, S. A. (2005). Determinants of public satisfaction with the police. Police Quarterly, 8, 279–297. Zavaleanu, D. (2010, November 25). Sondaj CSCI – 66 per cent dintre romani nu au incredere in politie. Cotidianul. Retrieved August 10, 2011, from http://www.cotidi anul.ro/sondaj-csci-66-dintre-romani-nuau-incredere-in-politie-130327/ Zvekic, U. (1996). The international crime (victim) survey: issues of comparative advantages and disadvantages. International Criminal Justice Review, 6, 1–21. Page 27