Andresen, M.A. (2012). Homicide in Lithuania. In M.C.A. Liem & W.A. Pridemore (Eds.), Handbook of European homicide research: Patterns, explanations, and country studies (pp. 437 – 449). New York, NY: Springer.
Lithuania is a relatively small Eastern European country in terms of both population and land area: approximately 3.4 million persons and 65 000 square kilometres. Lithuania is one of the Baltic States (Estonia, Latvia, and Lithuania) and is bordered by Russia, Belarus, Latvia, Poland, and the Baltic Sea. Within the Baltic States, however, Lithuania is the most populated and geographically largest country, divided into ten counties and sixty municipalities.
Classified as a transition economy, Lithuania declared independence from the Soviet Union in March 1990. This declaration was followed by significant economic adjustment that was classified as one of the worst decreases in standards of living at that time (Grennes, 1997; Jakubauskas, 2000). The Lithuanian economy recovered during the late-1990s with significant economic growth. And since the late-1990s, Lithuania has increasingly become part of the global economic community: associate member of the European Union (EU) in 1998, a member of the World Trade Organization (WTO) in 2001, and a member of the EU in 2004 (Ceccato, 2007; Jakubauskas, 2000). At this time, Lithuania is considered a liberal market democracy, with the majority of its international trade being with the EU.
Lithuania joining the EU is far from a non-trivial event. Over time, accession to the EU involves the standardization and/or synchronization of laws that govern international trade in goods and services, the free movement of factors of production (capital and labour), common (economic) development policies, and a common currency. Because Lithuania is at a significantly different level of economic development than the core of the EU there is the potential for significant economic, social, and legal change. As discussed below, accession to the EU has had an impact on homicide in Lithuania.
1.2. Country-specific details
Turning to social, demographic, and economic characteristics,1 Lithuania has a predominantly urban population (67 %) that has a large proportion of its population at working ages (15 – 64 years), 69.6 per cent; 14.2 per cent of the population is 0 – 14 years and 16.2 percent of the population if 65 years and over. The median age of the Lithuanian population is 39.7 years (37.1 years for males and 42.3 years for females). With a birth rate of 9.11 births per 1000 population, 11.18 deaths per 1000 population, and a negative net migration rate, Lithuania’s population is slowly decreasing.2 Dominantly composed of ethnic Lithuanians (83.4 %), the most prominent ethnic minorities are Polish (6.7 %) and Russian (6.3 %); similarly, Lithuania is dominantly composed of Roman Catholics (79 %), followed by Russian Orthodox (4.1 %) and Protestant (1.9 %). Literacy (those aged 15 and over who can read and write) is 99.6 percent and school life expectancy is 16 years—15 years for males and 17 years for females.
The rate of civilian private gun ownership in Lithuania is very low: 0.1 firearms per 100 persons. Of 179 countries, this makes Lithuania ranked 160. For a comparison, in the United States (ranked number 1) there are 88.8 firearms per 100 persons (Karp, 2007; UNODC, 2005, 2006, 2008).
Most criminogenic conditions within Lithuania have been improving in recent years, 2001 – 2008. Average monthly incomes, measured in constant Lithuanian litas, have doubled (982 to 2152); the unemployment rate in 2008 (5.8 percent) was almost one-third of the unemployment rate in 2001 (17.4 percent); and the percentage of the population with a university degree has increased significantly (11.6 to 16.1 percent). Not all aspects of Lithuanian society have fared so well, however.
The consumption of alcohol per capita in Lithuania, a common co-variate of homicide (Pridemore & Eckhardt, 2008; Rossow ,1996; Wells & Graham, 2003), has increased in recent years: a slow but steady increase from 10 (2001) to 11.4 litres of absolute alcohol (2008). Though such an increase may not appear to be problematic, this does represent a 14 percent increase in absolute alcohol consumption and a 57 percent increase in the per capita sales of alcoholic beverages.3 Moreover, there has been a sharp increase in alcohol-related mortality in Lithuania, 2001 – 2008. Alcohol-related mortality has increased for males and females in both urban and rural areas. Curiously, this increase appears to coincide with accession to the EU in 2004. Such an increase in alcohol-related mortality is not a surprise given the increase in alcohol consumption and does not bode well for crimes such as homicide. However, as shown below, homicide has had a downward trend over this same time period. Consequently, this increase in alcohol consumption and alcohol-related mortality is likely a result of increased (disposable) income that also increased sharply at the time of Lithuanian accession to the EU.
1.3. Previous studies of Lithuanian homicide
Aside from the research on recent trends in Lithuanian homicide discussed below, Värnik et al. (2003) considered Lithuanian homicide in an historical perspective in addition to investigating the effect of sociopolitical and economic conditions on homicide and suicide in the Baltic States. They undertook this study considering two sociopolitical periods: 1) 1970 – 1984, a period of stagnation under the Soviet Regime, and 2) a period of democratic reforms that began in 1985, then the move to a market economy in 1989, and subsequent stabilization, 1994 – 1998.
The authors found, as shown in Figure 1, that Lithuanian homicide in the first period (1970 – 1984) increased, though slowly. Immediately following the first democratic reforms (1985), the homicide rate fell to the level it was 10 years before. At the time Lithuania moved toward a market economy, the homicide rate rose sharply until 1994 when stabilization began to occur and the homicide rate decreased to its pre-1989 level. Needless to say, the homicide rate in Lithuania has been rather volatile over the past 40 years.
<Insert Figure 1 About Here>
2. Data source and measurement methodology used for this study
The data used for this study, including the alcohol-related statistics reported above, are all from Statistics Lithuania.4 The time line for most data is 2001 – 2008. This time line is used for reasons of data consistency. Data prior to 2001 are not used because of territorial boundary changes that occurred in 2000. Consequently, municipal level homicide data necessary for investigating the regional distribution of homicide are not consistently available (or reliable) pre-2001; after 2008, not all data are available.
Before the discussion turns toward homicide in Lithuania, a brief note is important regarding the measurement of homicide. The most common form of measuring homicide, or any crime for that matter, is a crime rate. The crime rate is calculated in order to control for the size of the underlying population: a larger country is expected to have more crime (homicide) simply because of its size. Though crime rates are known to have limitations (Boggs, 1965; Harries, 1981; Andresen & Jenion, 2010), particularly within the context of nationally measured homicide (Andresen et al., 2003), the standard crime rate is used in this chapter. This calculation is undertaken using the entire population of Lithuania as the population at risk.
In order to complement the use of a crime rate, the location quotient is also employed here. The location quotient was introduced to criminology by Brantingham and Brantingham (1993; 1995; 1998) and has been used by a number of scholars since (Rengert, 1996; Andresen, 2007, 2009a; McCord & Ratcliffe, 2007; Ratcliffe & Rengert, 2008). The location quotient is a geographical measure that may be used to calculate the “specialization” of an activity in an area that is within a larger region—Lithuanian municipalities within Lithuania, for example. The location quotient is calculated as follows:
where Cin is the count of crime i in municipality n, Ctn is the count of all crimes in municipality n, and N is the total number of municipalities. In the present context, the location quotient is a ratio of the percentage of homicide in a municipality relative to the percentage of homicide in all of Lithuania. If the location quotient is equal to one, the municipality has a proportional share of homicide; if the location quotient is greater than one, the municipality has a disproportionately larger share of homicide; and if the location quotient is less than one, the municipality has a disproportionately smaller share of homicide. Consequently, if a municipality has a location quotient of 1.20, that municipality has 20 percent more homicides than expected given the percentage of homicides in Lithuania—that municipality then “specializes” in homicide. Miller et al. (1991) provide the following classifications that are used in the maps below: very underrepresented areas, 0 ≤ LQ ≤ 0.70; moderately underrepresented areas, 0.70 < LQ ≤ 0.90; average represented areas, 0.90 < LQ ≤ 1.10; moderately overrepresented areas, 1.10 < LQ ≤ 1.30; and very overrepresented areas, LQ > 1.30.
Lastly, in the investigation of the regional distribution of Lithuanian homicide, two statistics of spatial autocorrelation are used to measure the presence (or lack thereof) of clustering in municipal homicide rates to answer the following question: do municipalities with similar homicide rates and location quotients cluster together? The first is Moran’s I that is a global measure of spatial autocorrelation that provides an indication of Lithuanian municipality homicide rates clustering. The second is local Moran’s I that is a local measure of spatial autocorrelation. Local Moran’s I provides an indication of each Lithuanian municipality’s homicide rate or location quotient being similar to that of its immediate neighbours (Anselin, 1995). This latter measurement of spatial autocorrelation is instructive because even if there is no global indication of clustering centred on one municipality, for example, there may still be spatial autocorrelation present around a small number of municipalities—this type of spatial autocorrelation simply gets “washed out” at the national level. All LISA calculations are performed using GeoDa 0.9.5i (http://geoda.uiuc.edu), a spatial statistical freeware package.5.
3. Epidemiology of Lithuanian homicide
3.1. Recent trends in Lithuanian homicide
According to data from the United Nations Development Programme,6 Lithuania has the 21st highest homicide in the world—only South Africa, Russia, and countries in Central and South America have higher homicide rates. The World Health Organization7 has also ranked countries based on their homicide rates and places Lithuania above much of Africa and the Caribbean, Mexico, Central America, South America, and a small number of countries in East Asia (Cambodia, North Korea, Myanmar, and the Philippines). Lastly, according to EUROSTAT,8 this trend continued up to 2008 in the European context. Currently, Lithuania has seven times the average homicide rate in the EU-27.9
Needless to say, the levels of homicide do not flatter Lithuania in an international context. However, it is possible that there is simply more crime in Lithuania, per capita, than in most other countries. Consequently, Lithuanian society may not be more violent than other countries, per se, Lithuania just has more of all crime. Unfortunately this is not the case. Based on EUROSTAT data, in Lithuania homicides accounted for 0.42 percent of all crimes in 2007. This is the highest for all EU-27 countries, doubling the next highest proportion of homicides, Latvia. Within the EU-15, homicides accounted for 0.02 percent of all crimes in 2007.
Though organized crime activities increased subsequent to Lithuania’s independence from the former Soviet Union and likely contributed to Lithuania’s high homicide rate, Lithuania is not the only country to have such issues (Juska et al., 2004). There is indeed something unique to Lithuania that has led to its international ranking in homicide. Needless to say, from an international perspective, Lithuania has both a high homicide rate and an overrepresentation of homicides.
Despite this negative reporting of homicide statistics, the Lithuanian homicide situation has been improving in recent years. From 1970 to the late 1980s there was a slight increasing trend that spiked in 1994 during the time of severe social, political, and economic adjustment after the collapse of the former Soviet Union. However, as shown in Figure 1, there is a clear downward trend in Lithuanian homicide, 1994 – 2008. There are two noteworthy spikes in 2000 and 2005, but the homicide rate is currently down by approximately one-third in the years immediately following its independence from the former Soviet Union. Moreover, the homicide rate calculations in Lithuania were changed in 2004. Prior to 2004, the homicide rate included homicide attempts. As such, one would expect there to be a drop in homicides at this point in the time series. Though there was a slight drop from 2003 to 2004, as just mentioned the homicide rate increased again in 2005. Consequently, the change in the definition of the homicide rate by Statistics Lithuania does not appear to have had much of an impact on the trend of Lithuanian homicide. This claim is substantiated through inspection of attempted homicide counts, post-2003: attempted homicides are a very small proportion of total homicides.
3.2. Regional distribution of homicide rates and location quotients
Considering the nation as a whole, Lithuania has a high homicide rate. However, this is not true for all regions within Lithuania. The regional (municipal) distribution of homicide rates (2007 – 2008) is shown in Figure 2a.10 Immediately evident is that there is significant variation in homicide rates across the Lithuanian landscape. During the 2001 – 2002 period, all but one municipality (Neringa municipality) had homicide rates greater than the EU average, but ten municipalities have homicide rates less than or equal to the average in the United States. Investigating the regional distribution of homicide rates in Lithuania reveals that not all Lithuanian municipalities have a particular problem with homicide.
There is little evidence of a spatial pattern to homicide in 2001 - 2002. There does appear to be greater rates of homicide in Kaunas and Vilnius counties, but the highest legend category on this map (15 – 25 homicides per 100 000) is represented in a number of municipalities that are not clustered together. Consequently, there is a need for the use of spatial statistics to identify the presence (or lack thereof) of homicide clustering at the municipal level.
The global Moran’s I for this map is statistically insignificant indicating that there is no spatial autocorrelation. Therefore, this result indicates that there is no global evidence of municipalities clustering together that have similar homicide rates. However, a local Moran’s I analysis reveals some minor clustering of high homicide rate municipalities in Kaunas and Vilnius counties.11 As such, there is some evidence to support clustering of homicide rates at the municipal level.
Figure 2a shows the average homicide rates from 2007 – 2008. Though there are fewer municipalities in the highest legend category, little has changed regarding the number of municipalities within each of the legend categories. Regarding the spatial pattern of homicide rates, it appears as though the municipalities with higher homicide rates (though not the highest) have moved north to Panevezys, Klaipeda, and Siauliai counties.
<Insert Figure 2 About Here>
Similar to the 2001 – 2002 results, Moran’s I is statistically insignificant. With a negative value of Moran’s I, this suggests that negative spatial autocorrelation is present to some degree. This claim is supported through a local Moran’s I analysis. However, statistically significant clustering is only present for seven municipalities.
Aside from showing that the regional distribution of homicide rates is not uniform across the Lithuanian landscape, nothing particularly interesting is present with the municipal-level homicide rates. The same cannot be said for the homicide location quotients.
Similar to the homicide rates for the same time period, the homicide location quotients appear to cluster in Kaunas and Vilnius counties. The global Moran’s I for 2001 – 2002 is statistically significant indicating the there is positive spatial autocorrelation: I = 0.11, p-value = 0.07. However, the local Moran’s I reveals very little spatial clustering for the homicide location quotient.
Despite a lack of spatial clustering, the homicide location quotients clearly show that a specialization in homicide is taking place. Moreover, that specialization is taking place in the same areas that have the highest homicide rates. And as would be expected, approximately one-half of the municipalities have homicides as a percentage of all crime that is below the national average.
Turning to the homicide location quotients for 2007 – 2008, Figure 2b, a very different pattern has emerged.12 Though there is little evidence for spatial clustering (I = -0.12, p-value = 0.11) and very few municipalities exhibiting local clustering when considering local Moran’s I, there appears to be a shift in the spatial pattern.13 Further spatial statistical testing is necessary to substantiate any such claims and is beyond the scope of this chapter, but homicide specialization appears to be moving toward most of the border municipalities. Additionally, less than one-third of Lithuanian municipalities have homicides as a percentage of all crime that is below the national average. This clearly shows the utility of using other measures than the crime rate to assess change.
Though homicide rates have been falling in Lithuania in recent years, this result does not bode well for those (new) municipalities with specialization in homicide. Though regional level data are not available to this author, an investigation into alcohol-related mortality and homicide in these municipalities is in order.
3.3. Homicide incident characteristics
There are a number of variables available that show trends in homicide incident characteristics including: juvenile homicides, male versus female victims of homicide, urban versus rural residence location of the victim14 (not necessarily if the homicide itself was urban or rural), the specific location of homicide, and the modus operandus. Statistics regarding these variables are reported in Table 1, for the period 2004 – 2008.15
Homicides committed by juveniles are a relatively small percentage of the total. This comes as no surprise as homicide typically has a later peak in its age-crime curve. Notable, however, is the increase in juvenile offenders. The three most recent years do exhibit a decrease in juvenile offenders, but all of these three years are greater than the first two years of available data. With such a short time series it is difficult to discuss any (changing) trends, but the appearance of an increase in homicides committed by juveniles definitely deserves future research attention as more data become available.
<Insert Table 1 About Here>
The percentage of male offenders charged with homicide is also no surprise. Most homicides, indeed most crimes, are committed by males (Boyd, 2000). With approximately 90 per cent of those charged for homicide in Lithuania being male, there is nothing unique regarding this characteristic. Similarly, males are most often the victim of a homicide. When comparing 2004 to 2008 data one may be tempted to state an increasing trend in this data series. However, the most recent four years of data are relatively constant.
Homicide in Lithuania is an urban phenomenon. Aside from aberrations in the 2005 data, homicides are consistently occurring in urban areas 60 per cent of the time. Additionally, when either males or females are victims of homicides, approximately 60 per cent of those victims reside in urban areas. Though caution must be taken with interpretations of these short time series, it does appear as though there are trends in the data: there appears to be a slight decrease in homicides involving urban males, and a moderate increase in homicides involving urban females.
The specific types of locations at which homicides occur have no variation over this short time frame so they are not shown in Table 1. The majority of homicides (51 %) occur within a residence, followed by public/open areas (37 %), nonresidential living areas such as cottages, hotels, etc. (5 %), other places (4 %), and public/interior areas such as hospitals, banks and restaurants (2 %).
Lastly, the modus operandus of homicides (Table 1) does exhibit some moderate variation over the time frame. In all years, the most common form of homicide is classified as “other”. Unfortunately no greater detail is available regarding this classification. Of those homicides with details provided, beatings are the most common, followed by stabbing, firearm, strangulation, torture16, and drowning. The low incidence of firearm homicide is also documented in other research (Karp, 2007; UNODC, 2005, 2006, 2008) and is likely a function of the low level of firearms ownership in Lithuania.
3.4. Homicide victim characteristics
The age breakdown of homicide victims in Lithuania is shown in Figure 3 for all victims, male victims, and female victims. Overall, females are victims of homicide in approximately 31 per cent of the incidents. With females having such high representation in victimization it is no surprise that all victims, male victims, and female victims have almost identical trends across the age categories. Clearly evident is that the per cent of homicide incidents increases with the age categories, the highest being those aged 40 – 59 years. However, this increase is partially artificial because the age categories in the data provided successively include more years.
<Insert Figure 3 About Here>
3.5. Homicide perpetrator characteristics
The age breakdown of homicide perpetrators in Lithuania is also shown in Figure 3 for all perpetrators, male perpetrators, and female perpetrators. Overall, females are only perpetrators of homicide in approximately 8 per cent of incidents. This result is consistent with previous research on this subject (Boyd, 2000). For perpetrators, females and males have a slightly different pattern. In the case of males, 78 percent of perpetrators are younger than 40 years, whereas only 64 percent of female perpetrators are younger than 40 years. Regardless, the majority of homicide perpetrators are less than 30 years of age. The same limitation regarding age categories is also present here for homicide perpetrators, but it is less pronounced because of the dominance of younger male perpetrators.
4. Explanations for homicide in Lithuania
The above discussion regarding homicide in Lithuania has been descriptive. Though descriptive statistics may be instructive, inferential analyses are necessary to uncover explanations for a phenomenon such as homicide. Though there is very little research on homicide in Lithuania, two recently published studies are reviewed below. The first covers those years almost immediately after Lithuania declared independence from the former Soviet Union, 1993 – 2000 (Ceccato, 2008), and the second covers more recent years assessing the impact of Lithuania’s accession to the EU, 2001 – 2006 (Andresen, 2010).
Similar to the discussion above, Ceccato (2008) finds that Lithuania has had relatively low homicide rates for the past 50 years when compared to Russia, but has had a significantly higher homicide rate than other European countries, particularly during the 1990s. Within the Baltic States, Estonia has historically had higher homicide rates than Lithuania. Though organized crime is often cited as being behind increases in crime, in general, and homicide, in particular, these only account for approximately 20 percent of homicides. Almost two-thirds of homicides involve friends or family members and alcohol. This does not bode well for Lithuanian society based on the increased alcohol consumption and alcohol-related mortality discussed above. Though homicide rates are falling (more than 30% during the 1990s) this may prove to be problematic in the future.
With regard to the regional distribution of homicide rates, Ceccato (2008) finds that Lithuanian homicide is higher in urban areas, particularly in the counties and municipalities mentioned above. These are the most densely populated areas and relatively affluent. The similarities in the regional distribution of homicide rates in Ceccato (2008) and presented above bodes well for ecological stability and inferential explanations for homicide in Lithuania at the municipal level.
In the inferential/explanatory analysis of homicide in Ceccato (2008), all Baltic States are analyzed collectively using a spatial statistical regression approach. The analysis is performed on 2000 data. Although the results presented by Ceccato (2008) are general for all of the Baltic States, specification tests were performed in order to identify statistically significant differences in the results. Only the results that are applicable to Lithuania are discussed here.
Ceccato (2008) puts forth a number of hypotheses to test with regard to homicide. She expects that reductions in population, signs of deprivation and social isolation, low levels of social care, higher divorce rates, and low voter turn-out lead to increases in homicide. These hypotheses are operationalized using the following set of explanatory variables: Divorce rate, deaths under one year of age per 1000 live births, hospital beds per 1000, gross domestic product per capita, foreign direct investment per capita, proportions of males 15 – 29, proportions of non-native population, natural (population) increase, net migration, voter turn-out, border regions, and population density.
The most important aspect of the analysis performed by Ceccato (2008) is the reliance on Western theories of crime to explain homicide in a transition economy. Many of these variables may be classified within social disorganization theory (Shaw and McKay, 1931, 1942) and are commonplace in ecological analyses of crime in Western contexts. Though not all variables are shown to be statistically significant, the explanatory power of these variables, discussed below, reveals that homicide in transition economies follows similar patterns as their Western counterparts—Pridemore (2005), Pridemore and Kim (2007) and Pridemore and Shkolnikov (2004) have also found that similar variables perform well in ecological analyses of homicide in transitional Russia.
The explanatory power of the homicide model still leaves much variation unaccounted for (R2 = 0.232), but a number of key results manifest from the analysis. First, the divorce rate is found to vary positively with homicide in Lithuania. Second, the increased presence of young males leads to increases in homicide. And third, the increased presence of non-native populations leads to increases in homicide.
Ceccato (2008) suggests that increases in divorce lead to increases in homicide because of the strain on families that resulted from the transition from a planned to a market economy. Recalling the proportion of homicides that involved family members and alcohol, this is a plausible explanation. However, it may also be the case that the standard social disorganization theory explanation applies: areas with higher divorce rates have more difficulty establishing a community that can repel crime—a socially organized area. This difficulty arises because more divorces usually mean more single-parent families and less time to interact with neighbours. The increased presence of young males is easily understood within the age-crime curve and the over-representation of males in criminal activity (Hirschi & Gottfredson, 1983; Steffensmeier & Allan, 1996). And the presence of non-native populations may be tied back to social disorganization through ethnic heterogeneity; areas with greater levels of ethnic heterogeneity have greater difficulty establishing social organization because of difficulties communicating or long-standing ethnic conflicts leading to higher levels of crime (homicide).
Lastly, Ceccato (2008) finds that the divorce rate, young male population, and non-native population are more powerful in explaining homicide than other variables such as foreign direct investment and geography (border regions). This bodes well for the use of Western theories in transition economies for understanding homicide.
Turning to the second study, Andresen (2010) focuses solely on Lithuanian municipalities. In his analysis, Andresen (2010) discusses homicide, rape, and robbery, but only the homicide context is discussed here. Methodologically, Andresen’s (2010) analysis takes a different approach than Ceccato (2008). Rather than analyzing a spatial cross-section, Andresen (2010) makes use of a panel data set: 59 Lithuanian municipalities across time, 2001 – 2006.17 Through the use of a panel data set and the appropriate statistical procedure (fixed effects estimation) a better indication of causality may be garnered from the results. When using a spatial cross-section, the research question one is implicitly asking is whether the spatial distribution of one variable coincides with the spatial distribution of another. This allows for the following type of question to be answered: is the homicide rate high in the same places that the divorce rate is high? Though instructive, the explanatory power (in terms of causality) is limited. In a panel data analysis the investigation involves asking: Do changes in one variable lead to corresponding changes in another variable?
Andresen’s (2010) research context is an investigation of the impact of accession to the EU on violent crime in Lithuania. Though the transition to membership in the EU is not as abrupt as the transition from a planned to a market economy, it is a transition nonetheless that is expected to disrupt aspects of society. Andresen (2010) takes an explicitly Western theoretical approach, invoking variable selection from social disorganization theory and routine activity theory (Cohen & Felson, 1979; Felson & Cohen ,1980, 1981): population density, average monthly income, relative income,18 divorce rate, unemployment rate, high school graduates, university graduates, and number of police per 1000 population.19
The explanatory power of the model for homicide is moderate (Adjusted R2 = 0.36). More importantly, variable retention and the corresponding interpretations for the homicide model are far superior to the rape and robbery results. He finds that increases in population density, income, unemployment, and university graduates lead to increases in the homicide rate. Though some of these relationships are not expected within the theories put forth, Andresen (2010) explains these results are relating to urban environments. As discussed above, and found in Ceccato (2008), Lithuanian homicide is an urban phenomenon. Population density is a proxy for urbanization and higher incomes and university graduates are more likely to be found in urban areas. He also finds that increases in relative income and divorce rates lead to decreases in the homicide rate. The negative relationship between relative income and homicide is an expected result, but not for the divorce rate—Ceccato (2008) found a positive relationship between divorce and homicide, as discussed above. However, Andresen (2010) explains this result appealing to levels of religiosity in Lithuania. In Lithuania, religiosity decreases with income; as such, municipalities with higher levels of income have higher levels of divorce and lower rates of homicide. This is consistent with social disorganization theory.
Turning to the results for the impact of accession to the EU, the trend in the homicide rate was decreasing before accession to the EU (see Figure 2) and decreased at a faster rate after accession to the EU. However, there was an abrupt upward shift in the homicide rate at the time Lithuania joined the EU similar to when Lithuania began the transition to a market economy (Värnik et al., 2003). Consequently, a faster decrease in the homicide rate post-EU is a good finding for Lithuanian society, but it comes at a cost of a sharp one-time increase in the homicide rate. This is interpreted as capturing changes in social organization and routine activities that are not captured in the list of explanatory variables.
Overall, Andresen (2010) is able to show that Western ecological theories of crime are able to be used successfully in the transition country of Lithuania, confirming the previous work of Ceccato (2008) on Lithuania and Pridemore (2005), Pridemore and Kim (2007) and Pridemore and Shkolnikov (2004) on transitional Russia. As such, homicide in Lithuanian municipalities may be (partially) explained using Western ecological theories of crime. This is an important finding because new or altered theoretical frameworks do not need to be generated to understand homicide in this transition economy.
5. Concluding remarks
Homicide in Lithuania is an interesting case study because it is so high in an international context. Moreover, because Lithuania is now a member of the EU that has some of the lowest homicide rates in the world, it will prove interesting how Lithuania adjusts to its now social, political, and economic reality. With so little information known about transition economies with regard to homicide and the availability of data from Statistics Lithuania, it is expected that Lithuania will become a research interest for many scholars in the future.
Andresen M.A. (2007). Location quotients, ambient populations, and the spatial analysis of crime
in Vancouver, Canada. Environment and Planning A, 39, 2423 – 2444.
Andresen, M.A. (2009a). Crime specialization across the Canadian provinces. Canadian Journal
of Criminology and Criminal Justice, 51, 31 – 53.
Andresen, M.A. (2009b). Crime in Lithuania: the impact of accession to the European Union.
European Journal of Criminology, 6, 337 – 360.
Andresen, M.A. (2010). The impact of accession to the European Union on violent crime in
Lithuania. European Sociological Review, in press. DOI:10.1093/esr/jcq039.
Andresen, M.A., Jenion, G.W., & Jenion, M.L. (2003). Conventional calculations of homicide
rates lead to an inaccurate reflection of Canadian trends. Canadian Journal of
Criminology and Criminal Justice, 45, 1 – 17.
Andresen, M.A., & Jenion, G.W. (2010). Ambient populations and the calculation of crime rates
and risk. Security Journal, 23, 114 – 133.
Anselin, L. (1995). Local indicators of spatial association – LISA. Geographical Analysis, 27,
93 – 115.
Boggs, S.L. (1965). Urban crime patterns. American Sociological Review, 30, 899 – 908.
Boyd, N. (2000). The beast within: Why men are violent. Vancouver, BC: Greystone Books.
Brantingham P.L., & Brantingham, P.J. (1993). Location quotients and crime hot spots in the
city. In C.R. Block & M. Dabdoub (Eds.) Workshop on crime analysis through computer
mapping, Proceedings (pp. 175 – 197). Chicago, IL: Criminal Justice Information
Brantingham, P.L., & Brantingham, P.J. (1995). Location quotients and crime hot spots in the
city. In C.R. Block, M. Dabdoub, & S. Fregly (Eds.) Crime analysis through computer
mapping (pp. 129 – 149). Washington, DC: Police Executive Research Forum.
Brantingham, P.L., & Brantingham, P.J. (1998). Mapping crime for analytic purposes: location
quotients, counts and rates. In D. Weisburd & T. McEwen (Eds.) Crime mapping and
crime prevention (pp. 263 – 288). Monsey, NY: Criminal Justice Press.
Ceccato, V. (2007). Crime dynamics at Lithuanian borders. European Journal of Criminology,
4, 131 – 160.
Ceccato, V. (2008). Expressive crimes in post-Socialist states of Estonia, Latvia, and Lithuania.
Journal of Scandinavian Studies in Criminology and Crime Prevention, 9, 2 – 30.
Cohen, L.E., & Felson, M. (1979). Social change and crime rate trends: a routine activity
approach. American Sociological Review, 44, 588–608.
Felson, M., & Cohen, L.E. (1980). Human ecology and crime: a routine activity approach.
Human Ecology, 8, 398 – 405.
Felson, M., & Cohen, L.E. (1981). Modeling crime trends: a criminal opportunity perspective.
Journal of Research in Crime and Delinquency, 18, 138 – 161.
Grennes, T. (1997). The economic transition in the Baltic countries. Journal of Baltic Studies,
28, 9 – 24.
Harries, K.D. (1981). Alternative denominators in conventional crime rates. In P.J. Brantingham
& P.L. Brantingham (Eds.), Environmental criminology (pp. 147 – 165). Prospect
Heights, IL: Waveland Press.
Hirschi, T., & Gottfredson, M. (1983). Age and the explanation of crime. American Journal of
Sociology, 89, 552 – 584.
Jakubauskas, E.B. (2000). The Lithuanian economy on the threshold of the new millennium.
Journal of Baltic Studies, 31, 127 – 135.
Juska, A., Johnstone, P., & Pozzuto, R. (2004). The changing character of criminality and
policing in post-socialist Lithuania: from fighting organized crime to policing marginal
populations? Crime, Law, & Social Change, 41, 161 – 177.
Karp, A. (2007). Completing the Count: Civilian firearms. Small Arms Survey 2007: Guns and
the City. Cambridge, UK: Cambridge University Press
McCord, E.S., & Ratcliffe, J.H. (2007). A micro-spatial analysis of the demographic and
criminogenic environment of drug markets in Philadelphia. Australian and New Zealand
Journal of Criminology, 40, 43 – 63.
Miller, M.M., Gibson, L.J., & Wright, N.G. (1991). Location quotient: a basic tool for economic
development studies. Economic Development Review, 9, 65 – 68.
Pridemore, W.A. (2005). Social structure and homicide is post-Soviet Russia. Social Science
Research, 34, 732 – 756.
Pridemore, W.A., & Eckhardt, K. (2008). A comparison of victim, offender, and event
characteristics of alcohol- and non-alcohol-related homicides. Journal of Research in
Crime and Delinquency, 45, 227 – 255.
Pridemore, W.A., & Kim, S-W. (2007). Socioeconomic change and homicide in a transitional
society, Sociological Quarterly, 48, 229 – 251.
Pridemore, W.A., & Shkolnikov, V.M. (2004). Education and marriage as protective factors
against homicide mortality: methodological and substantive findings from Moscow.
Journal of Quantitative Criminology, 20, 173 – 187.
Ratcliffe, J.H., & Rengert, G.F. (2008). Near repeat patterns in Philadelphia shootings. Security
Journal, 21, 58 – 76.
Rengert, G.F. (1996). The geography of illegal drugs. Boulder, CO: Westview Press.
Rossow, I. (1996). Alcohol-related violence: The impact of drinking pattern and drinking
context. Addiction, 91, 1651 – 1661.
Shaw, C.R., & McKay, H.D. (1931). Report on the causes of crime, volume II. Washington, DC:
Government Printing Office.
Shaw, C.R., & McKay, H.D. (1942). Juvenile delinquency and urban areas: A study of rates of
delinquency in relation to differential characteristics of local communities in American
cities. Chicago, IL: University of Chicago Press.
Steffensmeier, D., & Allan, E. (1996). Gender and crime: toward a gendered theory of female
offending. Annual Review of Sociology, 22, 459 – 487.
UNODC. (2008). Reported homicide numbers and rates, plus firearm-related homicide numbers
and rates per 100,000 population for 80 countries, 2005-06. Tenth United Nations Survey
of Crime Trends and Operations of Criminal Justice Systems (2005-2006). Vienna:
United Nations Office on Drugs and Crime.
UNODC. (2006). Reported Overall Homicide Numbers and Rates per 100,000 Population for 64
Countries, 2003-04. Ninth United Nations Survey of Crime Trends and Operations of
Criminal Justice Systems (2003 - 2004). Vienna: United Nations Office on Drugs and
UNODC. (2005). Recorded Homicide and Firearm Homicide in 53 Countries, Completed or
Attempted. Eighth United Nations Survey of Crime Trends and Operations of Criminal
Justice Systems (2001-2002). Vienna: United Nations Office on Drugs and Crime.
Värnik, A., Tooding, L., Palo, E., & Wasserman, D. (2003). Suicide and homicide: Durkheim’s
and Henry & Short’s theories tested on data from the Baltic States. Archives of Suicide
Research, 7, 51 – 59.
Wells, S., & Graham, K. (2003). Aggression involving alcohol: relationship to drinking patterns
and social context. Addiction, 98, 33 – 42.
Table 1. Homicide incident characteristics, 2004 – 2008
Gender and urban characteristics
Juvenile homicides, percent
Homicides, charged males, percent
Homicide, percent male victim
Homicide, percent urban
Homicide, percent male victim and urban
Homicide, percent female victim and urban
Modus operandus characteristics
Figure 1. Homicide in Lithuania, 1970 – 2008
Source. Värnik et al. (2003), Ceccato (2008), and Statistics Lithuania.
Figure 2. Regional variations in homicide, Lithuania, 2007 - 2008
b) Location quotient
Source. Statistics Lithuania.
Figure 3. Suspect and victim characteristics
Source. Statistics Lithuania.
Appendix for Springerlink Maps
Regional variations in homicide, Lithuania, 2001 – 2002, homicide rate per 100 000
Local Moran’s I, Lithuanian municipalities, 2001 – 2002, homicide rate per 100 000
Local Moran’s I, Lithuanian municipalities, 2007 – 2008, homicide rate per 100 000
Regional variations in homicide, Lithuania, 2001 – 2002, homicide location quotient
Local Moran’s I, Lithuanian municipalities, 2001 – 2002, homicide location quotient
Local Moran’s I, Lithuanian municipalities, 2007 – 2008, homicide location quotient