Homicide In Transition Economies; History Written In Blood.

The collapse of communism in Eastern Europe in 1989 and Soviet Union in 1991 are watershed events in world history. The demise of this system was cherished by many and a wave of democratization to replace the highly inefficient command economies followed soon after. Twenty-five years later the side effects of the transition processes that were initiated in the early 90's, characterised by immense suffering and pain in already fragile societies, are clearly visible. One such side effect was the immediate spike in criminal homicide rates prominent in almost every region that underwent this process. Was this a coincidence? What can we attribute this increase in criminal activity to? What impact did political and economic reforms have? What impact did cultural and demographic factors such as war and ethnic tensions have? I employ panel data analysis to answer these questions and handpick the most suitable control variables that explain homicide rates in transition economies, in order to build upon an already existing framework for such analysis.

My findings indicate that civil wars, where present, led to an inevitable increase in crime activity by desensitising the general population and perpetuating a culture of violence. Mass privatisation, a proxy of "shock therapy" applied to rapidly transform the economy also led to an undeniably large increase in homicide rates. This highlights the tradeoff between superior future economic performance and present human costs that policy makers were faced with. Trade liberalisation however, while it decreased criminal activity in the countries within the Commonwealth of Independent States (CIS) region such as Russia, Ukraine and Kazakhstan due to the reduction in black market activities, it promoted criminal activity in Central Eastern Europe (CEE) as globalization offered criminals more opportunities to exploit. Democracy, income inequality, ethnic tensions, inflation, GDP growth and alcohol consumption prove statistically significant for different countries within different regions, albeit to a different extent. Health expenditure on the other hand, a proxy for poverty proves insignificant for neither the CEE, nor the Baltic and CIS regions.

2. INTRODUCTION

At the beginning of 1989 the continent of Europe, dominated by the largest country on earth that no longer exists, the Union of Soviet Socialist Republics (USSR), embraced an economic system labeled as a command economy. "The aim to centralize the economic power follows from the belief that economic coordination by markets is inefficient and welfare is best maximized by centralized planning and administration" (Mickiewicz 2010). However, command economies were characterised by highly bureaucratic and extensive administrative structures that led to higher costs and distortion of information, a lack of economic efficiency derived from overproduction to meet the quotas set by the government and no profit motives for people as all economic parameters were decided by the planners. Ultimately the system was incompetent and it was the factors above, in addition to exaggerated military spending and public dissatisfaction that led to the political revolutions that started in mid-1989 in Central and Eastern Europe and Central Asia, a period of modern history otherwise known as "the post-communist transition".

Such transitions affected the life of the populations in the countries involved in many significant respects, a key aspect of which is population health. Safaei 92012) found that the countries of Central and Eastern Europe that had gone under immense political and socioeconomic restructuring after the collapse of communism in the 90s experienced sharp spikes in mortality rates. The notion that "transition had an immediate and largely adverse impact on health" is further reinforced by Figueras et al. (2004), yet the casual link between transition and increased mortality is widely debated by literature. Bobak et al. (2007) identifies income inequality and corruption as the source of poor health outcomes, Cockerham et al. (2002); Cockerham et al. (2006) found that people who favour political ideologies such as pro-socialism demonstrate less activity toward achieving health than antisocialists, Stuckler et. al (2009) discovered that economic reforms such as "mass privatization" in Eastern Europe and Soviet union were associated with a 12.8% increase in mortality rates, while Mckee and Shkolnikov (2001); Zatonski and Willett (2005) blame the life style factors such as the high prevalence of smoking, alcohol consumption and poor diet.

Additionally, Figueras et al. (2004) found increased mortality to be related to morbidity due to poorly organised health systems, while Franco et al.( 2004); Besley and Kudamatsu (2006) have drawn attention to lack of political culture and democracy, low levels of which were a symptom in a majority of countries that underwent this grueling process. Yet, it is the work of Leichter (1991) and Winston et al. (1999) best resonates with the focus of this dissertation, whose work centers on the connection between mortality and violence, an indirect result of disrespect for order and state control.

During the period of transition, countries of the former Soviet Union such as Russia faced a multitude of challenges related to crime, law and justice. This included drafting a new criminal code (Solomon 2005), corruption among the political elite (Coulloudon 1997; Wedel 2001), a police system with budget shortfalls and widespread corruption (Beck and Lee 2002), and consequently a judiciary distrusted by citizens (Huskey 1997). Equally, during the same time period of transition, the personal safety of the citizens of Eastern Europe were threatened by rising levels of criminal activity (Lotspeich, 1995; Savelsberg 1995; Zvekic 1998).

25 years after transition began, the assassination of Boris Nemtsov on the 27th of February 2015 is yet another bleak reminder of the still existing prospects of crime in post-communist countries like Russia, as it leaves liberals in fear of a new wave of violent repression. On March 7th The Economist (2015) quoted that "this postcard murder", referring to a symbolic image of Nemtsov's memorial procession with the cupolas of St Basil's church in the background, "marks the return of Russia's campaign of political violence from Ukraine to the homeland". Thus understanding crime using economic theory, or criminal homicide, the act of a human causing the death of another, using many forms including accidental or purposeful murder, has a profound academic significance.

On a social level, literature finds serious negative effects due to crime as well as fear of crime, particularly on psychological health (Morrall et al. 2010), on children's cognitive performance and further development (Sharkey 2010) and on the risks it poses on families' way of life as a sudden and uninvited intrusion in their lives that changes their meaning of existence, as they struggle to deal with the major transformations and challenges, that ultimately lead to them having a higher probability of developing sustained and dysfunctional psychological problems. (Thompson et al. 1998)

On an economic level, "homicide was estimated to decrease life expectancy by nearly 5 years"(Redelings et al. 2010), which is a measure of well-being in society, while also "acting like a tax on the entire economy: [as] it discourages domestic and foreign direct investments, it reduces firms` competitiveness, and reallocates resources creating uncertainty and inefficiency."(Detotto and Edoardo 2010)

The rest of the chapter is organised as follows. On section 3 I will present a literature review of homicide studies in transition, on section 4 I further analyse the dependent and independent variables involved in the study, in section 5 I discus the econometric model used to empirically asses my model and present the data, the results of which I discus and conclude in section 6.

3. LITERATURE REVIEW

In 2009, Stamatel (2009) used pooled time-series analyses of data from nine countries from 1990 to 2003 In East-Central Europe, to examine whether correlates of cross-national homicide variation tested with data from highly developed, predominantly Western Nations could also explain homicide rates in transition economies. His work built upon the four structural and cultural contexts that have been identified as correlates of cross-regional homicide in other countries by Gartner (1990); namely the economic context (distribution of economic resources), integrative context (integration of social networks and Institutions of social control), demographic context (the composition and activities of the population) and culture context (exposure to violent and legitimate violence). Stamatel concluded that within the economic context, the rate of homicides is negatively related to GDP per capita and not significantly related to income inequality, within the demographic context it's positively related to ethnic diversity and population density but negatively related to the percentage of young people between 15 and 25 years old and within the integrative context, its not significantly related to divorce rates, "a measure of intra-group cohesion [that] was anticipated to correspond to higher homicide rates". He discovered that homicides were also positively related to political violence within the culture context and additionally proved a correlation between lower homicides and successful economic reforms, as well as higher levels of democracy, as a result of a country's ability to recover quickly from the disruption of the end of the communist regime and to institute a new social order. Although the two latter factors are a subcategory of the economic and integration context, one must single them out into the economic reform and political reform context respectively, as they are rather unique features for transition economies.

LaFree and Tseloni (2006) reinforced the idea based on the conflict perspective, which predicts that violent crime rates will increase along with the brutalizing effects of the market economies that so far have universally accompanied democratization, as he found that "homicide rates in transitional democratic regimes were significantly higher by an estimated average of 54.4%". Nevertheless different academics use different explanatory variables within the four contexts outlined above to try and explain homicide rates in transition economies. For example according to Favarin (2013), the OLS model predicts that "a shift from the transition stage to a more solid democratic stage in Bulgaria, Romania and the Balkan countries, led to a reduction of homicide rates by 40.6 %", but LaFree and Tseloni (2006) conclude that "that during the second half of the twentieth century homicide rates gradually increased for full democracies", while "the average homicide rate for a hypothetical country of average percentage of population fifteen to twenty-four years old and average prosperity under autocracy is effectively zero". Nevertheless democracy facilitates a neutral stance towards ethnically diverse countries, thereby indirectly leading to a decrease in harassment, violence and ultimately murder, as emphasized by Ashforth (2005) who quotes that "differences of religion and culture ought to be treated as private matters to which governments should remain blind."

Other academics such as Pridemore and Kim (2007) have focused on the economic context as means of explaining homicide rates, as they find that "regions with greater increases in unemployment experienced greater increases in homicide rates" while "regions that privatized experienced smaller increases in homicide rates." Equally Prasad (2012) finds a negative correlation between trade liberalization and crime in transitioning India. While Stamatel (2009) concluded that within CEE countries income inequality was insignificant for homicide rates, other literature by Neopolitan (1996); LaFree (1999); Messner and Rosenfeld (1997) directly contradicts this. Ouimet (2012) also finds "that the strong bivariate relationship between the Gini coefficient of income inequality and homicide holds in a multivariate analysis for all countries, "but he limits this finding only for the subsample of medium to high Human Development Index (HDI) nations, thereby stating that inequality doesn't explain homicide rates for low HDI countries. Nevertheless according to the data from World Health Organisation, none of the countries in this study are ranked as countries with low HDI. According to (www.nationsonline.org) the low HDI threshold is 0.497. The lowest HDI reached by the countries in my sample was by Kyrgyzstan at an HDI level of 0.58 in year 2000 as demonstrated in graph 1.

Graph.1a UNDP Human Development Index HDI

The demographics context has also been under the scope with regard to homicide rates, as Kikuchi (2010) explores the spatial and temporal dimensions of crime in neighborhoods and finds that "racial homogeneity of the target neighborhood is most important when offending along with co-offenders" and Frühling et al. (2003) deduces that rising crime levels in other regions of the world such as Latin America for example, are attributable to ethnic diversity.

When identifying the main factors that lead to increased homicides in transition economies, it's important not to not leave important variables out of the equation, therefore violating one of the classical assumption of Ordinary Least Squares (OLS) regression theory (that the explanatory variables are independent of the error term) and hence skewing the results due to omitted variable bias. As a result one must control for all variables that might have caused a spike in homicide rates during the post-communist transition period. For example Ouimet (2012) uses the Gross National income (GNI), Gini coefficient and excess infant mortality, which is a proxy for poverty, to account for the economic context. To control for the demographic context he factors in the portion of 15-29 year olds and percentages of ethnicities as part of the population. To control for the culture context he includes a violent conflict dummy, and to account for the political context he uses a full democracy dummy and an authoritarian regime dummy. On the other hand, Pridemore and Kim (2007) use the ratio of the income received by the top 20% relative to the bottom 20% wage earners to account for inequality and therefore the economic context and the rate of heavy drinking per 100,000 deaths, the percentage of population living in cities with more than 100,000 residents and percentage of population aged 25-44 to account for the demographic context. To account for the integrative context, they use the rate of enrollment in college per 1000 residents and voter turnout as percentage of registered voters who voted in the 2000 Russian Presidential election. In order to ensure the model is as relevant and robust as possible, all different control variables used by different academics within each of the six categories examined so far will be thoroughly analysed. Crime rates increased during the period of transition, albeit at different rates for different countries. To the author's knowledge no empirical study distinguishes between the explanatory variables that might significant in the CEE, Baltic or CIS countries separately. What might explain the increase in homicide for one set of countries might not significantly correlate with the increase in crimes in another. This paper aims to bridge that gap by handpicking the best control variables for the set of countries involved in the study. Below, a table including all the contexts discussed above with all respective control variables for which either don't lack data or are believed to be worth testing based on supportive literature is devised. (See Table. 1 in Appendix 1) In the next section, each control variable is examined individually using collected data, and a comprehensive visual analysis is conducted to help devise the ultimate econometrics model that explains homicide trends in post-communist countries.