Systems of Violence and their Political Economy in Post-Conflict Situations.DOC

This research argues that post-conflict criminal violence in Guatemala and El Salvador is attributed to economic policies that weakened the state’s coercive and distributive capacities,

Systems of Violence and their Political Economy in Post-Conflict Situations

Nazih Richani

Associate ProfessorDirector of Latin American Studies Program

Political Science DepartmentKean Univesitynrichani@kean.edu

Work in Progress

March 2007

Abstract:

Conflict and post-conflict studies have paid inadequate attention to the political economy

of crime after the termination of civil wars, and the challenges that this poses to

reconstruction efforts Save few exceptions, the literature has largely overlooked the rising socio-economic and political costs of organized criminal violence, and its negative effects on creating conditions for sustainable development

The main goal of this paper is to fill an important gap in the literature by undertaking a comparative study of El Salvador’s, Guatemala’s, and Lebanon’s post-conflict

experiences It will attempt to answer why, in the post-conflict cases of El Salvador and Guatemala, there was an exponential increase in violence, while Lebanon witnessed a decline in crime compared to the level prior to the civil war This research argues that post-conflict criminal violence in Guatemala and El Salvador is attributed to economic policies that weakened the state’s coercive and distributive capacities, which in turn diminished the opportunity costs of crime This condition was further aggravated by contingencies Whereas, Lebanon’s economic policies and contingencies differed sharply from the other two cases under investigation, which may explain its low crime rates

This paper also demonstrates that criminal violence constitutes a serious threat as a return

to civil war In fact, in El Salvador and Guatemala, the costs of criminal violence have already exceeded those of their respective civil war

Finally, this paper adopts a systemic approach to explain the endurance of crime over time with its proper dynamic and political economy In this regard, one noticeable findingis: homicides rates are inversely proportional to the institutionalization of the system of violence

For the most part, researchers in post-conflict studies have focused on two

important sets of issues One set is primarily concerned with the conditions under which conflict may resume (Collier &Hoeffler, 2002, 2005; Sambinis & Doyle 2000; 2006) The other sets focus is security reform and policing as means to strengthen public

security in societies traumatized by civil wars (Call 2000; Call and Stanley, 2001)

Notwithstanding the important contributions of the mentioned literature, valuable

questions remain unanswered regarding the causes and permanence of high crime in conflict countries Questions such as: whether crime is causally linked to the economic policies adopted when civil wars terminate, or if crime is the continuation of the civil war

post-by other means High levels of post-conflict violence are underminingeconomic growth and threatening the consolidation ofdemocratic processes, both of which are core

contributors for successful reconstruction and subsequent political stability It also

demonstrates that criminal violence poses at least as serious a threat to development as a return to civil war (See table 2) A broader question asks whether the spike in crime is attributed to similar factors that govern crime in societies that did not witness civil wars This paper attempts to answer the first set of questions and provides insights that could help in answering the second

This paper undertakes a comparative analysisofcriminal violence in three

countries that have experienced civil wars: El Salvador, Guatemala, and Lebanon

Following the termination of their civil wars, El Salvador and Guatemalahave witnessed unprecedented levels of criminal violence, particularly homicides In 2005, El Salvador recorded a homicide rate of 54.7/100.000 and Guatemala 48.5/100.000 (La Prensagrafica,January 15, 2006; Prensa Libre March 3, 2006); however, in Lebanon the recorded rate

of post-civil war crime was as low as 2.2/100.000 in 2004, lower than the crime rate prior

to its civil war(Lebanon’s National Police, 2005) All three countries witnessed protractedcivil wars lasting until the early to mid 1990s El Salvador’s 12-year-long civil war wasbrought to a peaceful settlement in 1992; by then,75,000 deaths were attributed to the combat The Guatemalan conflict continued for 36 years and by the time a peace accord was signed in 1996, roughly 200,000 Guatemalans had been killed Lebanon’s 15-year civil war, which finally ended in 1990, was by far the bloodiest of the three; the conflict claimed 120,000 lives in a country whose population at the start of the war was 2.4 million, less than half the population of El Salvador and less than one-quarter the

population of Guatemala

Why compare El Salvador, Guatemala and Lebanon?

In this paper I follow the two main methods of comparative analysis: the “deep analogy” method as presented by Arend Lijphrat and Arthur Stinchcombe where the researcher chooses a few well matched cases; and the method of disagreement, where extreme cases are presented exhibiting analogous development The latter method is advocated by Adam Przeworski and Henry Teune Each method is supported by John Stuart Mill’s insights on comparative research.1

El Salvador, Guatemala, and Lebanon allow for the application of these two methods El Salvador and Guatemala are good candidates for the “deep analogy” method due to similar historical experiences regarding state formations, level of economic

development and the role violence since the 19th century Yet they demonstrate variations

in a dependent variable of post-conflict scenarios: crime rates.2 Three independent

1 My choice of the methodology was informed by the work of Fernando Lopez Alves, State Formation in Latin America 1810-1900 (Durham: Duke University Press, 2000), p.9.

2 Robert Holden, Armies without Nations: Public Violence and State Formation in Central America

1821-1960 (Oxford: Oxford University Press, 2006)

variables are explored in this research: state capacities and opportunity costs of both crime and contingencies

Lebanon and Guatemala represent the most different scenarios that allow for the application of the disagreement method El Salvador represents what Fernando Lopez-Alves called a “nexus case”, falling between the extremes of Lebanon and Guatemala.3

The three countries, however, are classified by UNDP as middle income, and they are ranked respectively according to the human development index: Lebanon 75; El Salvador104; and Guatemala 120 (Human Development Report 2002)

Understanding the dynamics of post-conflict violence and the factors that

contribute to it is imperative in creating effective policies to reduce crime and dismantle its tangible and intangible social and economic structures This paper argues that the high indices of crime during post-conflict periods are attributed to the formation of a system ofviolence, which has its own dynamics and political economy This system is formed whenthe following three variables intersect:

a) Low state capacities, measured in terms of the state’s coercive and distributive functions The coercive capacity is measured against the law

enforcementability of the state The percentage of criminals who remain

unpunished is a proxy indicator to measure the efficiency of law enforcement This indicator is complemented with an analysis of the justice system and

police/inhabitant ratio The distributive function of the state is measured by the level of the government’s social expenditures, including poverty reduction,

education, and health services State capacities in these two areas are fundamental

to effective governance

3 Ibid.

b) Low opportunity costs of crime Alternatives to crime are assessed

by evaluating average income obtained from crime versus minimum salary, levels

of unemployment, underemployment, and average education levels of the

criminal population

c) Contingencies Certain contingencies contribute to the formation of a system of violence, such as the repatriation of immigrants and narcotrafficking in El Salvador and Guatemala Alternatively, Lebanon experienced favorable post-civil warconditions that prevented the formation of a system of violence; a sharp decline in its population growth, an influx of investments, and highly skilled migrating labor

Hence, when low state capacities and low opportunity costs of crime intersect with contingencies that stress the state, the most likely outcome is a system in which the main interacting units are the specialists of violence (states’ agents, organized crime, political entrepreneurs, and private security) interact, forming a dynamic relationship and a political economy that institutionalizes and perpetuates violence Consequently, this paper defines the system of violence as a pattern of interaction sustained over time (say ten and more years)* and whose constituent units are: states’ agents, organized crime including gangs, political entrepreneurs and private security 4

However, due to the constraints of space, this paper focuses on two actors: state and organized crime and makes reference to private security and political entrepreneurs when deemed necessary

4 This definition is informed by the insights of Robert Jervis, System Effects: Complexity in Political and

Social Life (Princeton: Princeton University Press, 1998); and also draws on Nazih Richani, Systems of Violence: The Political Economy of War and Peace in Colombia (Albany: State University of New

York, 2002)

*The ten years or more figure is not totally arbitrary but is informed by the cases under investigation.

This paper is divided into five sections The first explores the role of former combatants in the subsequent violence that ensued Each case under investigation is discussed in a separate section (sections 2-4) The final section includes a

comparative analysis and preliminary conclusions

Post-conflict Violence and Former Combatants

The key question addressed in this section is whether criminal violence in El Salvador and Guatemala can be attributed to disgruntled former combatants being

recycled into the crime market, and whether the low crime rates in post-conflict can be credited to a successful process of demobilization El Salvador was more successful compared to Lebanon in demobilizing its combatants; however, the crime rate in El Salvador increased, while in Lebanon it decreased This confirms that post-conflict criminal violence corresponds to distinct causes and dynamics that cannot be attributed toformer combatants (See following sections) In Guatemala, crime is a more complex phenomenon because it is related to the combination of former (either purged or

demobilized) and active members of the state security, together with youth gangs conflict violence in Guatemala is characterized by former and current elements of the military protection-racket state

Post-In El Salvador, the average age of the criminal population ranges between 15 and

21 years, indicating that they were born between the years of 1977 and 1983, and were not combatants during the civil war (Cruz, 1998) Former members of the insurgency were integrated into a newly formed police force. 5 In 1994 and 1996, the number of

5 In El Salvador 20,000 soldiers were demobilized and 10, 490 combatants from the FMLN (7,500 of those chose to engage in agriculture, 1600 chose to start a small business And 699 chose to continue their education See Centro de Estudios Internacionales, Hablan los Desmobilizados de Guerra: Nicaragua, El Salvador and Mozambique (Managua: CEI,1995), p76.

homicides spiked, largely attributed to reprisals related to the conflict; however, the violence between 1997 and 2005 was subject to different causalities and dynamics

About two third of Lebanon’s 30,000 demobilized fighters were incorporated into the armed forces; the remainder (circa 10,000) joined private security companies,

emigrated, or returned to their professions.6 This partial demobilization was not coupled

with the removal of light weapons nor was it complete Hezbollah, a military structure

with several thousand fighters, deployed along the border with Israel, retained its

weapons Nevertheless, the homicide rate in Lebanon (2.2/100,000) is amongst the lowest

in the world (Lebanese Police)

Former military and ex-Civil Defense Patrullas De Autodefensas Civil (known by the acronym PACs) have become among the main perpetrators of criminal violence in Guatemala during the post-conflict period A number of former generals removed after the failed coup d’etat of 1993 constituted the nucleus of a criminal organization known as

GS.7 This group was primarily involved in narcotrafficking and at least five retired military officers were allegedly involved in this group, two of whom played leading roles.8 In fact, in November 2005, the Minister of Defense, General Carlos Aldana

Villanueva, and the chief of the antinarcotics division of the police were removed from office because of blatant links between the military and criminal elements, including the construction of a landing strip in an army base financed by organized crime (Prensa Libre, November 21, 2005)

6 A number of combatants were not full timers or if they were, they did for a limited time which allowed to keep their jobs or practice their professions See Elizabeth Picard, Prospects For Lebanon: The

Demobilization of the Lebanese Militias (London: Center for Lebanese Studies, 1999), p.29.

7 David Keen, Demobilising Guatemala Working Paper no 37 (London: London School of Economics, 2003), p.16 It is reported that the retired generals Lius Francisco Menaldo and Manuel Callejas are two

key figures accused of narcotraffking see Prensa Libre, November 22, 2005.

8 Ibid.

The insurgent army in Guatemala did not exceed more than 1,500 combatants when the peace agreement was signed Hence, their reintegration into civilian life was relatively seamless, however, the real problem stemmed from an inflated security

apparatus that was largely geared toward containing the insurgency For example, the PAC, which was formed between 1978 and 1982 and banned after the accords, had about 900,000 rural men between the ages of fifteen and sixty years (representing nearly 80%

of the male rural indigenous population)(Bruneau, 2004:21) When it was dissolved in

1995, about 375,000 were still in the patrols These patrols did not receive any salaries but acquired significant power in their villages, in turn allowing them to improve their social status and material conditions Some of these ex-patrol members may have also been involved in criminal activity (Interviews, Guatemala City, 2005)

Demobilization in Guatemala led to links between the military and criminal elements, whereas in El Salvador the successful demobilization of civil war combatants produced a different outcome in terms of criminal violence; alternatively, the partial success of demobilization in Lebanon did not impact crime rates What explains the differences between the outcomes of demobilization in these three countries? In the following sections, I argue that the answer lies in the dynamic interplay of the three variables: the post-conflict states’ capacities, low opportunity costs of crime, and specific contingencies

The Paradox of El Salvador

Between 2000 and 2004, El Salvador witnessed 10,000 homicides, making it one of the most violent places in the world In the 12 years that followed the 1992

peace accord, an estimated 49,317 individuals were killed, which is equivalent to 2/3

of the 75,000 deaths recorded during its civil war (UNDP, El Salvador, 2005)

The main contributors to violence in El Salvador are gangs and

narcotraffickers The first part of this section examines the genesis of the gang

problem in El Salvador, which in 2004 was responsible for the 60% of the country’s homicides (Ribando, 2005)

Gangs

Following the peace accord, El Salvador’s economy was transformed into a service economy and largely dependent on the remittances of Salvadorans in the Diaspora In 2004, the remittances reached 16.2% of the GDP.9 Between 1979 and

1991, thousands of Salvadorians left their country, heading mainly to the U.S These numbers convey the scope of the exodus: in 1980, only 94,000 Salvadoran-born people resided in the U.S., by 1984 this number quadrupled to 400,000, and by 2000 the number was 1.2 million or 40% of the population of El Salvador in 1980 (Hayden,2004: 202)

The immigrants were unskilled, mostly peasants, with low levels of education and speaking little, if any, English Naturally, they were at a great disadvantage in the U.S labor market This disadvantage, coupled with the U.S economy’s limited

capacity to absorb such a large pool of unskilled labor, particularly for those settling

in Los Angeles, led to high levels of unemployment among the Salvadoran

immigrants, making them a fertile recruitment pool for the inner city crime economy (Hayden, 2004) In the 1980s hundreds of these immigrants were drawn into the then

flourishing gangs such as the 18th street gang ( Hayden 2004; Davis, 1992).

9 http://www.bcr.gob.sv/publicaciones/main_comunicado12005.html

The 18 th street gang offered protection in tough neighborhoods to some

Salvadoran and Guatemalan immigrants who shared similar backgrounds in terms of class, language and professional skills The new Salvadoran immigrants started

calling themselves “la Mara Loca” (the “crazy group”)

Mexican immigrants established the 18th street gang in the 1940s and

continue to constitute its core Initially, the relationship between the La Mara Loca

and the 18th street was good, but the introduction of drug economy stirred their ethnic

differences leading to gang warfare in competition for a larger share of the drug market The ramifications of this warfare and the respective narratives that each gang constructed about its identity came to form an integral part of today’s gang violence and dynamics in El Salvador and Guatemala alongside the U.S, Honduras, and

Mexico

In this process of identity building, la Mara Loca started calling itself Mara

Salvatruchas Stoners: Salvatruchas in recognition of their country origins and Stoners

since rock music was their favorite beat Later they dropped the Stoners from their

name and retained Mara Salvatrucha or MS (Hayden 2004:204) The MS expanded

its membership to where the Salvadoran youth lived in Los Angeles, and later on to San Francisco, New York, Long Island, Reno, and Texas, and in most recent times Washington, DC, and Virginia The FBI estimates that the MS has about 10, 000 members in the U.S and is active in smuggling immigrants along the Mexico-U.S border.10 Since the mid 1990s, the MS and the 18 Street gangs have exacerbated the

crises of governance in El Salvador and Guatemala, contributing to the formation of the systems of violence in their respective countries

10 Se http://www.pbs.org/newshour/extra/features/july-dec05/gangs_10-05.htmle

Evolution of State Capacity

Between 1960 and 1974, the homicide rate in El Salvador ranged between 23.6 and 32.0/ 100,000 The average for this same period was 28.9/ 100.000,11 and the trend was relatively stable After 1992, however, the homicide rates reached unprecedented levels of 138.2/100,000 (in 1994) and 138.9/100,000 (in 1995); the rates declined

considerably after 1995 but never returned to pre-war levels The sharp increase in 1994 and 1995 is attributed to the aftershocks of the civil war, which included political

assassinations, vendettas, and other motivations related to the civil war In 2004, the Attorney General Office (Fiscalia General) reported 3,997 homicides in a total population

of 6,704,943 people, which put the rate at 59.6/100,000; more than double the average of the prewar period (see figure 1) slightly decreasing to 55/100000 in 2005.12

Figure 1 Rates of homicides/100.000 in El Salvador

1960-2005

Rate of homicides/100.000

0 50 100 150

r

Sources: (Steiner, 1999; Fiscalia General, 2005, PNC 2006)

The state capacity is a two-pronged concept demonstrating the ability of a state to enforce the rule of law and the effectiveness of its distributive functions.13 Authoritarian

11 Roberto Steiner, “Criminalidad en El Salvador Diagnostico y Recomendaciones de Politica” (El

Salvador: FUSADES, February, 1999).

12 This ratio is my calculations based on the number of homicides reported by the Fiscalia General, 2005 The 2005 ratio is based on the PNC reports.

13 The introduction of the state capacity in this paper was in part informed by Charles Tilly’s

conceptualization See Charles Tilly, The Politics of Collective Violence (Cambridge: Cambridge

University Press, 2003), chapter 2

regimes dominated El Salvador for most of the twentieth century, and the military played

a significant role in maintaining the rule of law and controlling politics This arrangementwas cemented during the civil war by the support of the established oligarchy and the U.S Nonetheless, by the late 1980s, theUS changed its policies and played a significant role in fundamental changes to the country’s elite composition; in particular, helping to dismantle the existing oligarchy and undoing its alliance with the armed forces This opened the door for the emergence of a new hegemonic force: the agro-business and financial-service elite, represented then by an emerging leadership of the Arena Party, which came to power in the mid 1980s (Lungo, 1990:21-5; Robinson, 2003:95) This prepared the country for the introduction of electoral democracy coupled with neoliberal economic policies which began in 1988, and were pursued in earnest in the 1990s Thesechanges diminished state capacities

Distributive capacity

The neoliberal economic policies led to the reduction of the state’s role in the economy and in social spending, both of which had a significant impact on access to social services for the poor, including health and education Public health expenditures remained at 3.6% of the GDP between 1990 and 2002 (UNDP, 2005:309: see table 3) The chronic budgetary crisis of the public health and education systems led to the

bifurcation into public and private, where parts of these services became subjected to the laws of exchange value and profit (Robinson, 2003:245) In this bifurcation, the poor received a less favorable deal than those who could afford paying for their health care Between 1998 and 2000, the proportion of public expenditures on education decreased from 16.6% to 13.4%, in terms of percentage of the GDP, it decreased from 3.1%

between 1985 and 1987 to 2.5% between 1995 and 1997 and slightly increasing between

2000 and 2002 to 2.9% (Human Development Report, 2003; 2002; 2005; see table 3)

The breakdown of the peasant economy led to a sharp increase in the urban population In 1980, the urban population was 40% of the total population: by 1990, it was 47% and by 2001, it had reached 61.3%, an increase of more than 20% in two decades (Human Developmemt Report 2003: Seligson 200) At the same time, the

contribution of the rural economy declined from 38 % of the GNP in 1980 to 13% in

1998 (World Bank 2000) Most of this rapid urbanization process and the decline of the agrarian economy was accelerated after the end of the civil war, particularly after the introduction of trade liberalization alongside the other two pillars of the Washington Consensus: fiscal austerity and privatization

The emerging service sector was unable to absorb the growing number of urban laborers The increase in the population density of the marginal barrios, in a country characterized as the most densely populated in the mainland of the Americas (838

persons/ square mile), stressed the state’s capacities beyond its ability to deliver services Fourteen years after the peace accord, these barrios continue to have poor sanitation services, unsafe drinking water, unpaved roads, scarce entertainment resources for the youth, high unemployment, and more important, high underemployment rates Although

an unemployment rate of 7% is relatively manageable by world standards,

underemployment is very high; it reached a 17% in 2002, up from a 15.4% in 2001 About 85.5 % of the underemployed work forty or more hours per week and receive less than the minimum wage (Tolentino & Lopez, 2003) The minimum salary is about

$150/month

This background of a weakened state capacity coupled with underemployment and low minimum wage salaries led to two interrelated developments that can explain thenew trend in criminal violence The first development was the deportation of thousands

of illegal Salvadoran immigrants from the U.S after the signing of the peace accord The second, an interrelated development, was the introduction of cocaine, and its cheaper derivative crack, into the Salvadoran market

Repatriation and State Capacities

The U.S forcibly repatriated an estimated 130,000 immigrants between 1993 and

2003 Of them, 43,000 had a criminal record and many were members of the MS and the

18th street gangs These two gangs soon spread with ease in the marginal barrios of San

Salvador and other departments such as Sonsonate, and Santa Ana, two of the most violent departments in the country Active gang members reported to the author that they were very often able organize and recruit without detection, capitalizing on the

limitations of the state and the poor conditions in the marginal areas (Interviews, 2005) Some of these repatriated immigrants were knowledgeable about the crime trade, use of weapons, and more importantly, the production and distribution of crack They also maintained their communication with gang leaders in Los Angeles and elsewhere in the U.S Beginning in 2005, the Department of HomelandSecurity launched a national antigang campaign that led to the arrest and deportation of more than 1,260 gang

members, of which 293 were Salvadoran, further fueling gang activity in El Salvador (Prensa Grafica, September 2, 2005)

Although a few youth gangs existed in El Salvador in the 1980s, gang

membership and scope of activities skyrocketed by the late 1990s Roughly 35.6% of the

country’s population is under the age of 15, and is vulnerable to recruitment in gangs (Human Development Report 2002; Hayden 2004) Estimates of the combined

membership in the two main gangs range from 10,500 to 38,000 (Cruz, 2002; Miranda, 2005) If we accept the 10,500 figure, this is almost 2/3 the size of the police force and nearly the size of the army If the other criminal organizations are combined, a more sizable crime community emerges, in turn has exacerbating the weakness of the coercive capabilities of the state

Since the peace accord was signed, the police force was reduced by 6,092 to 16,000 agents This reduction occurred when the country was undergoing an accelerated process of “deruralization” due to the breakdown of the peasant economy as mentioned before.14 In El Salvador the ratio is about 238 police per 100,000 people, on the lower end according to a Rand report, particularly if we discount the number of police

dedicated to administrative work In fact, in August 2005 the government enlisted 1,000 army personnel to join forces with the police in an effort to fight crime

The homicide units of both the Fiscalia and the police have complained that their 10-year-old equipment and limited number of personnel make it impossible to cope with the growing number of homicides Consequently, the collection of forensic evidence is compromised, which in turn makes it more difficult to press charges, let alone convict (Mayorga, 2005)

Low Opportunity Costs and Deterrence: The “Suicide Bomber Syndrome”

The expected utility model of economics and rational choice concepts has

dominated crime theories since the late 1960s The utility theory posits that individuals

14 According to a Rand study the level of police should be at least 200/ 100,000 inhabitants 5 years after the termination of any conflict Jones, et al 2005: Establishing Law and Order after Conflict, Rand I think this figure could be disputed in the context of El Salvador where the rate is 238/100,000 throughout the 1990-

2005 period and criminal violence remained high by interntaional and regional standards

take actions (whether lawful or unlawful) that maximize their payoff and minimize their costs (Becker, 1968; Reynolds, 1995; Heineke, 1978; Posner, 1986) One derivative of rational choice that is relevant to this study is deterrence theory, which contends that potential criminals calculate the expected legal punishment that they may receive if caught Consequently, criminal behavior depends on the results of this calculation; if the expected pain of punishment offsets the potential reward, criminal behavior may be discouraged (Gibbs, 1975)

In El Salvador as well as Guatemala, the state’s limited controlling capacity supports the premises of deterrence and rational choice The apparent lack of legal deterrence is evident considering the high levels of crimes that go unpunished (more than 90% in El Salvador), which in turn increases the incentive for criminal behavior This high level of impunity is coupled with another important element that exacerbates crime: prison is no deterrence I interviewed and surveyed 20 active gang members shortly after

the government of Antonio Saca introduced the “Super Mano Dura” (Extra Tough on

Crime policy) against gangs and criminal organizations, resulting in hundreds of arrests.(Mayorga, 2005) Eighteen of these members reported that they do not fear the police or being sent to prison.15 They explained that life outside prison is tougher than inside it

Contrary to what the government anticipated, the campaign has not proved to be adeterrent; in fact, the crime rate has increased in the months following the launch of this policy Between January and July 2005, for example, homicides increased by 39.3% fromtheir 2004 levels, reaching the highest level since 1999.16 In order for a deterrence to

15 Those interviewed and surveyed are from 6 municipalities: Santa Tecla (2); Quezaltapeque (2); Opico (2), Colon (4); Apopa (4); and San Salvador, colonia Dina (6) All were males; 10 are members of the MS and

10 are members of the 18 gang

16 This is based on Police Records See Prensa Grafica, August 2, 2005; see also Sociedad Sin Violencia, October 28, 2005 http://www.violenciaelsalvador.org.sv/modules.php?

occur, the state must enhance its coercive capacity to establish law and order, at the same time offering its citizens the opportunity of a quality life meeting minimum standards of health, education, and social welfare Lower costs of crime result in effective deterrence The 20 individuals interviewed and surveyed were young, poor, had an average schooling

of 6 years, had low skill levels, and bleak assessments of their future Eight out of 10 did not think they had much to lose if they continued their criminal behavior, even if that meant dying in the process (Smutt & Miranda, 1998) Some even laughed when asked about their future; they share a strong sense of fatalism due to the high levels of death rates among gang members.17 Such a disheartening forecast affects the manner in which these individuals calculate the opportunity costs of their violent behavior There is a low appreciation of life given their class backgrounds, lack of education and skills, and their current living conditions, which pose what I call the “suicide bomber syndrome,” where deterrence and the instrument of dissuasion are outweighed by the despair or commitment

distributive functions, as is the case of El Salvador and more so in Guatemala

The increase in human rights violations of a citizens’ right to due process is an indicators of the return to authoritarian practices This was manifested by the anti-gang

Act of 2003 that was repealed in 2005 after strong national and international opposition, currently new efforts are being made by the government to revive certain aspects of the Act Moreover, in the last few years several assassinations were reported in prisons, the latest was in June 2005 which led to the death of 35 prisoners, some of whom were gang members, two prison guards were implicated in the incident.18

Finally, in the assessment of opportunity costs it is important to note the level of economic returns that sustain criminal behavior On average, a gang member could generate between $300 and $1000/month, far exceeding the minimum salary ($150) in a scarce labor market and rampant underemployment.19 This income is primarily generated from protection rents, which they extract from local businesses, residents, distributors of merchandise, and drivers of public transportation, and proceeds from selling crack

cocaine These findings complement those of a study establishing that the average family income in areas where there is gang activity is $257 (Cruz, 2004:307)

Narcotrafficking and the Surge of Crime

The emergence of narcotrafficking has created additional incentives for criminal behavior given the relative ease with which the crime organizations operate Prior to the 1990s, El Salvador was not a transit point for narcotrafficking Salvadoran anti narcotics police reported that cocaine began to appear in the local market in the mid 1990s This was attributed to the use of cocaine as a form of payment to the local groups who helped

in the trafficking (Interview, Police Antinarcotics Unit, 2005) Cocaine generally

18 www.amnestyusa.org/countries/elsalvador/document.do?id=ar&yr=2005

19 This finding was supplemented with information provided by informants that also are working with gang members If we consider on average that a gang member generates as a minimum a $300/month and we have about 10,000 gang members these will amount to 300.000/month x 12=36.000.000/year are mostly transfer and drug trade trade This would plausibly constitute the gang economy although that I think this is

a conservative estimate In contrast, in Guatemala, the estimated average income of a gang member is in the order of $405/month See Guatemala’s section.

purchased by two different markets: the high-end local consumers who can afford the market price and the low-income consumers of crack The drop in the price of one

kilogram of cocaine in the local market (from $25,000 in 1995 to $10,000-15,000 in 2005; Interview Anti Narcotic Unit PNC, January 26, 2005) meant that the supply was increasing despite the antidrug campaigns spearheaded by the U.S in the Andean Region, the primary source of the world supply

The consequences of the emergence of the drug economy were multifaceted Narcotraffickers hire young people to act as drug sellers some of whom are members of

MS and 18 street gangs Some in turn become consumers A drug-runner explained to me

that selling crack provides him an income that he cannot expect from working in the formal economy, given his low skill level and lack of education When asked about the possibilities of being caught, he did not seem concerned The income from the drug economy depends on where a given group/ individual stands in the division of labor Gang members are mostly at the lower end and make anywhere between $15/ week/per person to more than $600/week/per person (interviews; questionnaire) Entrepreneurs hold a higher position in the trafficking chain; with more successful entrepreneurs

creating relationships with drug traffickers in both Mexico and the U.S.20

The total size of the Salvador drug market and the possible economic gains for those at the top are difficult to estimate However, the amount of cocaine and marijuana decommissioned between 1995-2005 is revealing: the amount decommissioned increasedfrom 14 kg of cocaine in 1990 to 87 kg in 1995 to 2,703 kg in 2004; whereas the

decommissioning of marijuana increased from 85 kg in 1990 to 181 kg in 1995 and to

20 This information is based on my structured interviews with active gang members in El Salvador (See footnote 10).

402 kg in 2004 (Policia Nacional Civil, 2005) This means that either the police are becoming more efficient in their antinarcotics operations or the amount of trafficking is

on the rise The U.S Department of State is inclined to the latter view (International Narcotic Strategy Report 2005)

The police estimates that the market value of the decommissioned drugs in 2004 was on the order of $36 million This is a fraction of the real size of the drug economy According to police sources, the El Salvador drug market is very modest by regional standards, yet large enough to increase the incentive for criminal behavior The drug economy led to an increased number of homicides related to turf wars, contract violationsand control of routes between narcotraffickers and competing gangs

The Economy of the System of Violence

The full integration of El Salvador into global markets was spearheaded by the agro-industrial-financial service elite, which realized that ending the civil war was

imperative to realizing this integration (Robinson 2003: 101) The peace accord, however,did not make drastic changes; it secured the socio-economic order and introduced only some modification to the institutional order to incorporate the FMLN into the political system as a price to abandon its armed struggle (Robinson 2003: 101)

The incorporation of El Salvador into global markets depended on the

implementation of fiscal austerity, trade liberalization, and privatization, which weakened

or overwhelmed the state’s distributive and coercive capacities, a classic shortcoming identified by Stiglitz (Stiglitz, 2003) This made El Salvador’s political economy more vulnerable than before to contingencies such as changes in the U.S policies toward Salvadoran illegal immigrants (including those with criminal records) and the increased

activity of narcotrafficking since the mid 1990s These two contingencies together with the grave economic conditions of the rural economy and low opportunity costs of crime overwhelmed the state’s capacities, contributing to the formation of a system of violence

El Salvador’s system of violence serves as a mechanism of income redistribution through protection rents and robberies In 2003, it is estimated that about $307 million were transferred through robberies (PNUD, El Salvador, and 2005:58) The system of violence also provides business opportunities to the security commodity to which about

$411 million is invested by private enterprises and individuals (PNUD, El Salvador, and 2005:58) Private security companies’ employ about 20,000 agents, whose salaries amount to 0.5% of the GDP (Interviews) The combined income generated fromthesystem of violence amounted to 4.8% of the GDP in 2003 This does not include the drug economy and other illicit activities On the other hand, the costs in terms of loss of life, medical and institutional costs, decreases in investments, and losses in consumption and labor time amounted to 6.6% of the GDP in 2003

The formation and consolidation of the system of violence was exacerbated by the deportation of gang members from the U.S and the emergence of narcotrafficking Political entrepreneurs such as the ruling Arena party have capitalized on heightened insecurity to strengthen their grip on state institutions Antonio Saca’s “tough on crime” platform helped him win the 2004 presidential elections, and his cabinet is faithfully working to keep that pledge However, considering that the indices of crime since the introduction of the “Super Mano Dura” (the Extra heavy Hand) policy have increased, one can conclude that this policy has been unsuccessful This increase in violence could

be attributed to the relationship between opportunity costs and deterrence With a scarcity

of the carrots required by the orthodox neoliberal economic model, crime becomes to some the only viable option in the absence of minimum social safety nets Consequently, the interplay of the fight against crime and criminal behavior and the overall political economy that their dialectics create is consolidating a system of violence that threatens the fragile democracy in El Salvador The alarming increase in human rights violations and against citizens’ rights to due process attests to the degradation of the

democratization process (Human Rights Watch Report 2004) Hence, it is possible that further consolidation of the system of violence could generate authoritarian civilian strands under the pretext of fighting crime and maintaining order and security

Finally, this system of violence has become more than the sum-total of its parts I have attempted to explain the conditions under which this system was established, its genesis, and its consolidation Further research is needed to determine how this system acts upon its units and how it impacts factors such as regime type, criminal behavior, policy making, allocation and distribution of resources, opportunity costs, and its

relationship with the international political economy But here, suffice to say that

homicides rates are in function of the system’s institutionalization as discussed in the following section

Guatemala State Capacity in Guatemala following the Civil War

Like El Salvador, crime in Guatemala particularly homicides is due to the development of a dual power structure in which the state- private security nexus and organized crime have established an uneasy coexistence leading to the formation and consolidation of a system of violence In the capital, for example, entire neighborhoods

are virtually off limits to the state forces According to Arnoldo Illasco, the vice president

of Ebano (one of the most important private security company in Guatemala), “almost 60% of the capital city is under the control of organized crime” (Illasco, 2005) Other officials and business groups’ representatives interviewed confirm this assessment

If the state is incapable of extending its authority in the capital city, it certainly cannot project its power onto the periphery In many departments the state authority is meager, and illegal forces such as narcotraffickers can easily circumvent it This duality

of power contributed to the emergence of a system of violence that is more consolidated than the one in El Salvador Consequently, this degree of institutional development of thesystem may explain why Guatemala has lesser homicides than El Salvador since in this latter the specialists of violence (state agents, gangs and narcotraffickers) are yet to establish hegemony or stable balance of power

In the 1970s, the murder rate was 21/ 100.000;21 it rose to 61/100,000 in the 1990s (Steiner, 1999:12) (figure 2) In spite the great reporting discrepancies among the different state’s agencies regarding crime statistics22 which makes it difficult to draw definite conclusions in this comparative study, a trend is noted that during the post-conflict period murder rates were about three fold higher than the rates of the early 1970s

In 2005, the homicides rates reached 48.5/100.000 This latter calculation is based on the 5,338 reported murders in a population of 11 million (Morales, 2006) Sergio Morales,

21 InterAmerican Commission on Human Right by Manuel Bermudez, Central America: Gang Violence and Anti-gang Death Squads, InterPress Service News Agency Sept 6, 2005: http://ipsnews.net/news.asp? idnews=30163 ).

22 Accuracy of data in the case of Guatemala is questionable; there is no single source that is authoritative The homicides data of the Public Ministry (Ministerio Publico) are significantly higher than the National police For the years 2000 and 2001, the ministry reported 7,232 and 7,573 respectively; whereas the police reported for the year 2000 2167 homicides and 2304 for the 2001 The discrepancy is much higher for the year 2004: the police reported 4,507 homicides, the Justice Department 5,103, and the Public Ministry 11,236 (see El Periodico May 6, 2005 P.3) According to these figures, and on the base of a population of 11 million, the homicides rates are respectively 40/100,000 (Police); 46/100.000 (Justice Department); and finally 102/100,000 (Public Ministry)

Human Rights’ Ombudsman, revealed that the number of murders in 2005 exceeded the average 4000 homicides recorded during the civil war (Morales 2006)

Figure 2 Rates of Homicides in Guatemala 1970-2005

Homicides Rates/100.000

0 20 40 60 80

year

Source: (Steiner 1999:12; Morales, 2006)

The following sections assess the state’s distributive and coercive capacities and explore how the reductions in state’s capacities have facilitated the emergence and the subsequent consolidation of its system of violence

Guatemala allocated only 1.4% of its GDP to public education, whereas El Salvador

23 Between 1995 and 2000 the fertility rates of El Salvador was 3.2 percent; Guatemala 4.9 percent and Lebanon 2.3 (Human Development Report 2002).

spent 2.9% of its GDP and Lebanon 2.7% A December 2000 MINIQUA report found that

51 percent of the population are children under age 18 and that 83 percent of these

children live in poverty (see table 3) More importantly, less than half the school age population has access to primary education, and only 3 out of 10 students who begin primary education complete it.24

In terms of concentration of income, Guatemala has a 59.9 Gini index, which is higher than El Salvador (with GINI of 53.2) and is the highest in the Western Hemisphere(Human Development Report, 2005:272-3) There is no indication that the governments that came to power after 1996 attempted to introduce policies to redress this social

imbalance These governments failed to introduce a tax system that could enhance the state capacity and help it to increase social spending in priority areas At 8% of GDP, Guatemala had the lowest tax ratio in the hemisphere (Robinson, 2003: 114) The accord called for an increase to 12% by 2000, which is still below the average of developing countries (14%) The issue of taxation is still to be resolved; the government gets 80% of its tax returns on consumption and only 20% from income and wealth

The income distribution gap between the richest 10% and lowest 10% is the highest in Guatemala, followed by El Salvador and Lebanon Although these social indicators per se do not explain why Guatemala has such high indices of violence, it provides the context

Coercive capacities

Similar to El Salvador, the state in Guatemala largely depended on authoritarian policies for social control Since the 1954 coup d’etat the alliance between the military

24 www.state.gove/g/drl/rls/hrrpt/2001/wha/8344

and the landed oligarchy provided the social base for a regime that excluded the majority

of the population from the political process and capital accumulation The 1996 peace accord and in contrast with El Salvador and Lebanon, was between a considerably weak insurgent force and key sectors of the dominant classes whose interests were affected by the nuance of an ongoing low intensity conflict that hindered Guatemala’s full integrationinto global capital

In retrospect, the balance of forces produced a post-conflict political situation thatneither affected the socio-economic order nor did it provide the muscles for a meaningful opposition that could hold governments accountable (Robinson, 2003:113) This lack of accountability, in part, explains how some old regime security structures-such the

infamous paramilitary known by its acronym PACS- banned under the accord were resurrected and also how some of the new regime’s security apparatus came to form part

of organized crime A January 2002 Hemisphere Initiatives (HI) report describes the nexus between the remnants of the pre-accord security organizations and the criminal organizations that evolved after 1996 It explains that “The merging of clandestine

counter-insurgent structures that developed during the internal conflict with organized criminal gangs” resulted in an increase in crime At the same time “there was evidence that some of the perpetrators of crimes had been part of clandestine paramilitary

structures during the internal conflict (Hidden Power, 2003)

In this mode, Guatemala represents an extreme case where some members of the old regime coercive structures alongside ones that are still active in the police and army came to constitute the core of new crime groups During the civil war these security organizations acted as a protection racket and a brutal instrument of social control, in the

post-conflict an important segment of these institution resorted to crime to compensate for the losses that the peace agreements entailed: budget cuts and size reduction

(Rodgers, 2004; Friedman 2003) Consequently, this has weakened the state and has generated a governance crisis A condition that was best described by the President of the powerful chamber of Commercial, Agricultural, Industrial and financial institutions (CACIF), Roberto Ardon, he said “the state in Guatemala is like vapor it exists but is not there” (Ardon, 2005)

The weakened state in its turn provided additional incentives for criminal

behavior, particularly narcotrafficking, which found in Guatemala a safe haven Ardon in

my question about his assessment of criminal organization and activities, he said that criminal organizations pass through three phases, the formation phase, the parasitical phase in which they extract rents in the form of extortion and from providing security, and finally when organized crime is successful it forms a symbiotic relationship with the state (Ardon, 2005) He did not specify whether Guatemala is in its third phase yet, but it

is plausible that given the high levels of corruption, particularly inside the ranks of the political elite and the law enforcement agencies, Guatemala may be approaching the thirdphase

In February 2007 by the Ministry of Government (Ministerio de Governacion) confirms such conclusion The Ministry’s report claims that narcotraffickers exercise dominant political role in the departments of Jutiapa, Jalapa, Zacapa, Izabal, Petén and San Marcos, and they have penetrated also the political establishment in Iscuintla, Quicheand Alta Verapaz (Prensa Libre February 19, 2007).25

25 A similar is Colombia where narcotraffickers have also made similar political inroads into congress, and exercising political control of several governorships in several departments (Richani, 2007).

The coercive organs suffer from two intertwined problems One is rampant corruption and secondly a relative modest size given the size of the territory and the population With a total police force of about 19,000, and a population of 11 million, that puts the police/civilian ration at 170/100,000, which is among the lowest in our three cases and perhaps among the lowest in the world Lebanon, for example, with a

population of 3.5 million has a 15,000 strong police force, that is a ratio of 428/100,000 and territory of just 10,452 square km which makes the coverage of 0.6 square km/police,whereas Guatemala’s is 108,890 that almost 10 times the size of Lebanon with a

population of 11 million The police ratio/ territory are of 5.7 squares Km/per police officer El Salvador police/civilian ratio is 238/100,000 and police/square km is of 1.2

km square/officer One can expect lower crime in El Salvador than Guatemala since it hasmore police/habitants/Km but this is not case, homicides remains higher in El Salvador than Guatemala

Table 1 Police Force and Ratio per Inhabitant and Area Coverage

is of course much more expensive and the use of its services is limited to the middle and

26 In this respect it is important to note the increase numbers of linchings carried out by people against common criminals This practice is a response to state’s failure in protecting the lives and properties of citizens It is becoming a kind of a “parallel justice” that forms part of a system of violence.

upper income groups This state of affairs created a great demand for private security, which soon was supplied by 400 legal and illegal companies with a total work force that ranges between 60,000 and 80,000 employees (Illascos 2005; Ardon 2005; Parra, 2005) This number exceeds the combined number of police and army available This created a bifurcated system of security one private that caters for those who can afford it and the other public characterized with incompetence and corruption, which in turn reinforces theimmediate demands for more private security

Given that the lower class cannot afford such a service and to guarantee some level of normalcy in their neighborhoods, they are forced to pay protection rents to organized criminal groups The protection tax ranges from $2/day to $10 depending on the kind of business (Interviews; Prensa Libre, May 11 2005:6) In this mode, criminal violence (real or perceived) contributes and reinforces a duality of security providers (on one side the state- private security companies, and on the other side organized crime) which demonstrates an existing power structure between these actors More important, this duality is reinforced by the demand for security of two groups of consumers, the haves and the have-nots

Organized Crime and Opportunity Costs

Notwithstanding the poor social conditions of the majority of the Guatemalan population and the limitations of the state coercive and distributive functions, these factors alone do not explain why the system of violence emerged and reached such a level

of institutionalization “Specialists in violence” acted upon such conditions There are three main clusters of organized criminal groups operating in Guatemala similar to El Salvador and in order of importance these are: narcotraffickers, gangs and smaller groups

made up of 34 organizations specialized in activities such as car theft, kidnapping, home robberies, and assaults (El Periodico, May 6, 2005:3) Within the first group there are three or four main cartels structures that control the routes of narcotrafficking and with connections with their counterparts in Colombia, Mexico and the U.S The MS and the 18th Street gangs constitute the second cluster.

Narcotrafficking

The geostrategic location of Guatemala and the inability of its state to broadcast authority contributed to the growth of an elaborate drug economy with its organizational structures Narcotraficking organizations with relative ease control the cocaine routes in strategically located departments such as Peten, Izabal, Jutiapa, Santa Rosa and Escuintla.These departments constitute an arch that covers the entry /exit points from the Pacific coast, El Salvador, Honduras, Caribbean Sea, to Mexico In these places the rule of law

is a rare m commodity and a number of drug cartels exercise authority co-opting the localpolice and other law enforcement agencies It is no surprise that these same areas witness the highest number of homicides where exercising control on routes and incompliance with contracts are mostly settled violently

The state-crime relationship, or “symbiosis”, referred to before, has reached new heights in the last few years The U.S Drug Enforcement Agency (DEA) described the situation as follows “intelligence indicated that Guatemalan law enforcement officials profited from stolen drug loads; offered protection for traffickers and their illicit cargo