(Re)Shaping the Neoliberal Leviathans: the Politics of Penality and Welfare in Argentina, Chile and Peru

Going beyond general depictions of convergence in penal and welfare policies targeted to manage the urban poor and the flexibilized working class in neoliberal regimes of Latin America, I address the political causes behind the distinct penal and welfare policies developed in Argentina, Chile and Peru since their neoliberal turn. To explain the initial differences of penal and welfare regimes among these cases and their evolution, I integrate Harvey’s and Wacquant’s perspectives on state policies under neoliberalism and complement them with an analysis of local political conditions and processes, following Portes. The differences in penal and welfare policies in each country result initially from the political regime that governed the transition to neoliberalism – authoritarian, semi-authoritarian or democratic. Their consolidation or modification resulted from the organizational features (technocratic or neo-populist) of the political parties that governed the aftermath of transition to neoliberalism and from the different reactions of marginalized urban sectors to neoliberal adjustments and policies. Resumen: (Re)modelando a los leviatanes neoliberales: la politica penal y social en Argentina, Chile y Peru Mas alla de las descripciones generales de la convergencia entre las politicas penales y sociales dirigidas a los pobres urbanos y a la clase obrera flexibilizada en los regimenes neoliberales de Latinoamerica, yo abordo las causas politicas que se esconden tras las distintas politicas penales y sociales elaboradas en Argentina, Chile y Peru desde su giro neoliberal. Para explicar las diferencias iniciales entre los regimenes penales y sociales de estos paises y su evolucion, integro las perspectivas de Harvey y Wacquant sobre las politicas estatales bajo el neoliberalismo y las complemento con un analisis de las condiciones y los procesos politicos locales, de acuerdo con Portes. Las diferencias entre las politicas penales y sociales en cada pais radican inicialmente en el regimen politico que goberno en la transicion hacia el neoliberalismo – autoritario, semiautoritario o democratico. Su consolidacion o modificacion fue el resultado de las caracteristicas organizativas (tecnocratas o neopopulistas) de los partidos politicos que gobernaron despues de la transicion al neoliberalismo y de las distintas reacciones de los sectores urbanos marginados ante los ajustes y las politicas neoliberales. Palabras clave: neoliberalismo, penal, politicas de bienestar, partidos, movilizacion urbana, Peru, Argentina, Chile.


Making sense of a variety of neoliberal leviathans
Despite a growing literature that confirms the development within neoliberalism of an expanded penal state, with harsher policing, criminal sanctions and higher imprisonment rates (Godoy 2005;Auyero 2010;Iturralde 2010;Muller 2012) along a focalized welfare (Kaufman and Segura-Ubiergo 2001;Svampa 2005) targeting the urban poor, few attempts have been made to explain variations in penal and welfare statecraft within Latin America as related to broader political transformations.A rich literature remains focused on the micro-level issues of criminal justice or welfare reform.This is surprising given the striking differences between neoliberal policies in the welfare and penal fronts in different countries, both at the beginning of the transition to neoliberalism and after.If we take Argentina, Chile and Peru, we find that in Chile, at the moment of imposition of the model (1975), poverty relief and unemployment compensation was initially extended, targeting the destitute and the elder, but later on it was limited, being strictly means-tested and demanding recipients to looks for jobs and retraining (Vergara 1994;Repetto 2001).Housing for the urban poor became subsidiary to the market (Vergara 1994, 241) and welfare has remained small since then (Raczynski 2002).Regarding the penal state, it grew at the moment of transition and continued expanding in democratic times, when it was thoroughly reformed.The police adopted new deployment models and community policing (Carabineros 2005); criminal courts were reformed and prisons were partly privatized and implemented in an extensive parole system.
In neoliberal Argentina, imposed by democratic government, we find first a reduced welfare and a mildly repressive penal state.Right after the transition to neoliberalism in 1990, unemployment compensation was minimal (Etchemendy 2004b), with President Menem (1989-1999) implementing a 'fragmented and dispersed' welfare (Cerruti and Grimson 2004), abandoning the notion of welfare as a right (Lo Vuolo and Barbeito 1998).This policy was reversed in the late 1990s as poor relief and unemployment insurance expanded and today aims to have 'a general reach to all unemployed' (Cerutti and Grimson 2004, 23).This clientelistic welfare coexists with subsidized work programmes managed by local governments and, until recently, organizations of the unemployed.
Regarding coercion, the penal state has expanded somewhat, but remains authoritarian and corrupt without having been reformed.Police reforms toward greater legality, community policing styles and managerial improvement started in early 1990, but were reversed a decade later (Ungar 2002), preserving its authoritarian style (Brinks 2008, 44-45); criminal pro-cedure reforms have been limited, with judges controlled by the executive branch, and the prison system remains overpopulated, continuing to be a 'ware-housing' prison.
Finally, Peru began, like Argentina with a very limited welfare, but contrary to Argentina, stayed small.After two decades the Peruvian welfare system is a small means-tested system of assistance and poverty relief through a variety of targeted programmes that 'develops within clientelistic formats, oriented to satisfy beneficiaries that have turned into powerful interest groups, not necessarily the extreme poor' (Tanaka and Trivelli 2002).On the penal front, contrary to Argentina but closer to Chile, neoliberalism was inaugurated with utmost violence by Fujimori in Peru.After Fujimori the police grew, but post-Fujimori reforms toward legality and managerial modernization were discontinued.By contrast, 'the state translated the strategy used against terrorism to fight urban crime' (Frühling 2003; see also Basombrio Iglesias 2006), while municipal surveillance forces and neighbourhood watchmen organizations (serenazgos) became providers of security.Criminal courts have been also expanded but remain unreformed (Hammergren 1998), and although prisons have expanded in infrastructure, they are overpopulated and practically controlled by inmates (Bermúdez Tapia 2007).How can we understand these very different penal and welfare policies in these three neoliberal regimes?
To account for these divergent paths I integrate Harvey and Wacquant's views about the expansion and harshening of the penal state and the reduction of welfare policies under neoliberalism, and combine them with Portes' research programme on neoliberalism in Latin-America focused on the analysis of: a) 'the character of the state apparatus and its relations with different sectors of civil society'; b) 'the density of social networks among its less privileged sectors'; and c) the 'unexpected responses by the groups most directly affected' (Portes 1997, 254).

Toward understanding variations in neoliberal penal/welfare policies in the periphery
Both Harvey (2005) and Wacquant (2009) detect a tendency toward an expanded penal system and a reduced subsidiary welfare targeting the urban poor and informalized working class in countries that turn to neoliberal regimes.Each of them emphasizes different aspects that lead to these changes, specific targets of these policies and ways in which penal and welfare policies interrelate.For Harvey, neoliberalism is a 'political project to reestablish conditions of capital accumulation and class power ' (2005, 19), centred on promoting the power of financial capital, reducing state expendi-tures, privatization, and flexible labour markets.In the state, this means there are not only economic, but also specific penal and welfare policies.The main modality of repression is 'coercive legislation and policing tactics [...] to disperse or repress collective forms of oppositions to corporate power ' (2005, 77 ) combined with 'discarded workers and marginalized populations' controlled through increased surveillance, policing and incarceration.Within his perspective 'in developing countries, where the opposition to accumulation by dispossession may be the stronger, the neoliberal state quickly assumes the form of active repression, even to the point of lowlevel warfare against oppositional movements, criminalized as terrorists or drug dealers ' (2005, 165).These variations in coercion coexist with a 'minimized' and focalized instead of universal welfare, 'emphasizing individual responsibility ' (2005, 76).These penal and welfare measures operate alternatively.
For Wacquant (2008), as neoliberalism is 'the bureaucratic response of political elites to the mutation of wage work ' (2009, 5), he privileges political causes and effects.For him 'neoliberalism entails the enlargement and exaltation of the penal sector of the bureaucratic field, so that the state may check the social reverberations caused by the diffusion of social insecurity in the lower rungs of the class and ethnic hierarchy as well as assuage popular discontent over the dereliction of its traditional economic and social duties ' (2008, 12).Regarding targets, the penal state in neoliberal regimes, both in the core and periphery, mainly targets not the organized working classes, but the 'the insecure fractions of the post-industrial proletariat via the wedding of social and penal policy at the bottom of the polarized class structure ' (2008, 18).Through penal expansion, authorities 'reassert the authority of Leviathan so as to bolster the evaporating legitimacy of elected officials ' (2008, 14).Penal expansion is accompanied by retracting welfare turning into a disciplinary workfare.Penal control and welfare intervention, in turn, are complementary, producing a 'double regulation of poverty by the joint action of punitive welfare-turned-workfare and a diligent and belligerent penal bureaucracy' (2008,3).If we combine Wacquant and Harvey, we are to expect that penal repression will vary from control of individual delinquency, through police surveillance of organized opposition all the way up to military repression of organized opposition, depending on the level of organized class opposition to neoliberalism.Following Wacquant, we take into account political determinants for the changes in the penal and welfare state within the neoliberal project.Combining them is useful: with Wacquant we correct Harvey's class-based lenses, paying attention to autonomous political interests of legitimation and the need to secure followers among the excluded working classes and urban poor, in particular in situations of low levels of organized opposition.With Harvey, we correct Wacquant's top-down approach introducing contestation.
To understand the initial policies we need to look at the types of political regime that engaged in neoliberal structural adjustment -authoritarian, semi-authoritarian or democratic -and the degree of organized opposition they faced.The consolidation or reversal of these initial policies are conditioned and impacted by the organizational transformation of political parties that govern the transition and/or its aftermath and the systemic or antisystemic nature of mobilization of the urban marginalized groups.
The initial extension of welfare and the intensity of repression during and immediately after the transition varies according to the democratic or authoritarian nature of the regime that imposes it and the acquiescence or resistance of organized groups to neoliberal restructuring.According to Etchemendy (2004a) in transitions to neoliberalism under authoritarian regimes and with organized labour opposition, welfare and poor relief, paradoxically, tends to expand, even if it is focalized, whereas in democratic transitions to neoliberalism, that incorporate organized labour within the reform coalition, welfare remains limited, restricted to the participants of the coalition (Etchemendy, 2004a).Regarding penal coercion, I expect that repression increases in authoritarian regimes, but following Harvey, that it does even more when they face organized opposition to neoliberalism (2005,165) and labour is left outside the reform coalition.Conversely in conditions of lesser open resistance, following Wacquant, I expect a greater role for police and judicial targeting of disorganized marginalized groups in lieu of military repression.
I am not just interested in the initial policies.I also want to shed light on their evolution.These policies change or consolidate depending on the transformations of leftist or 'popular' political parties and to the impacts of urban poor grassroots mobilizations in relation to neoliberal adjustments.
The consolidation of the initial neoliberal welfare and penal policies adopted depends not just on the ideological orientation of the centrepopular parties that governed the transition to neoliberalism and its aftermath, but most importantly on the ways in which parties produce votes.In parties that produce votes mainly through a distant and mass-mediated modality of reaching out to the voters from the flexibilized working classes and the urban poor, as in Chile, poverty assistance and penal policies are reinforced by the neoliberal ideology, and their implementation is less affected by political interference.In governments controlled by neo-populist and clientelistic parties as in Peronism in Argentina and Fujimorismo and APRA in Peru, one observes less ideological purity and a less coherent implementation of these neoliberal policies.In these clientelistic parties, focalized-welfare is incorporated within clientelistic networks targeting the urban poor, providing the bases for the neo-populist reproduction of party allegiance, or allowing for the re-appropriation of such assistance to build new bases of popular contestation.The mass-mediated or clientelistic modes of reaching to possible votes also impact penal policies in different ways.
Mainly mass-mediated parties are more prone to promote a discourse of penal toughness and severity ('penal populism') but also of administrative efficacy.Penal populism gets combined with a managerial modernization, improving and expanding police, courts and prisons efficiency and accountability, capturing the allegiance of the less privileged sectors via providing them security.In parties-turned-clientelistic political machines, 'penal populism' coexists instead with political uses of bureaucratic corruption: using profits from corruption in clientelistic networks and intraparty competitions, and to neutralize alternative popular mobilization or repress urban marginals.The 'technocratization' and core reliance on marketing techniques by Chilean alliance of leftist parties -'Concertación' (Roberts 1998a), the mutation from a union-based party into a political machine based on clientelism of the Partido Justicialista (PJ) in Argentina (Auyero 2000;Levitsky 2005), and the neo-populism of Fujimori, Toledo and APRA in Peru (Weyland 1996;Roberts 1998;Barr 2003) all impacted the initial welfare and penal policies.
The initial policies were changed or reinforced by urban poor grassroots mobilizations.Following studies on urban grassroots mobilization from Chile (Garreton 1994;Paley 2001), Argentina (Svampa 2005;Garay 2007) and Peru (Roberts 2005) that show that marginal urban civil society reacts in different ways to the imposition of the neoliberal market and state restructuration, I distinguish between movements that are consistent with the new policies or disruptive of them.The resistance or backing of subaltern civil society to the commoditization of the labour market, the diminished welfare and the expansion and harshening of the penal branch have important effects on consolidating or changing those neoliberal welfare and penal policies implemented.

Chile: authoritarian transition, mass-mediated parties and reinforcing mobilization
We can trace the extended but focalized welfare and expanded repressive penal apparatus we observe today in Chile back to its authoritarian modality of transition to neoliberalism and the opposition it experienced.In this authoritarian transition in a context of left-wing mobilization, neoliberalism started with 'parties, labour federations and peasant associations dissolved [and] factories, universities and shantytowns occupied by the security forces, who killed thousands of left-wing activists who resisted the coup or were detained in custody' (Roberts 1998a).With the radicalized left quickly destroyed, the military, aided by a new breed of technocrats (Silva 1991;Etchemendy 2004b, 355-60) privatized and liberalized the economy and flexibilized industrial relations (Etchemendy 2004a).Since the late 1970s, the government and its technocrats imposed new neoliberal welfare schemes.From there they provided only to the atomized urban poor an 'extensive net of unemployment compensation that absorbed a significant part of the jobless [51 per cent at the 1983 peak of unemployment].However, more important than direct unemployment subsidies were the employment programs financed by the state' (Etchemendy, 2004, 281).Besides unemployment and poor relief, the government promoted private ownership of houses and aid for extremely needy shantytown dwellers, letting the market provide for those who could pay -the great majority.
In the coercive front of the state we observe that the military, intelligence and the police expanded their budgets and operative targets under the doctrine of national security, squashing political opposition and urban disorders (Colectivo Memoria Histórica Domingo Cañas 2005;Policzer 2008).In 1985, the government built maximum-security prisons to house terrorists, and organized a new parole service and a system of alternative sanctions imported directly from the U.S. (Chile 1980).The expanded police combined with a reinforced justice increased the number of imprisoned since 1980, duplicating it during the 1990s.While crime rates decreased by the end of the 1980s, the prison population and those under penal surveillance started increasing, since 1984 from 116 up to 412 per 100,000 today (Dammert 2008).
The structural adjustment policies brought changes in the economy and in the Santiago urban landscape consequential for ulterior political party dynamics and urban civil society reactions.Chile diversified its economy, exploiting natural resources, investing in human capital, incorporating large groups within the flexibilized labour market, reducing both extreme and relative poverty, and reducing unemployment (Silva 1993).In turn, Santia-go experienced a massive immigration from the central valley that followed the reacquisition of expropriated lands and the ulterior rationalization of the agrarian industry.It also saw a sustained investment in infrastructure in the working class and the 'campamentos' or urban sectors occupied during the Allende years.The informal urban sector turned into a community of property owners, who would distinguish themselves from the more precarious and poverty and crime ridden sectors occupied by the recently migrated to the southern part of the city (Sabattini and Wormald 2003).
The ideological and organizational changes in traditional centre-left political parties that governed the aftermath of the transition to neoliberal structural adjustment reinforced the initial options.The structural adjustments under dictatorship produce a dissociation of parties and labour and led traditional centre left parties to adopt pro-market ideas along new organizational modalities of producing votes.In democracy in Chile the Concertación coalition continued with the neoliberal developmental approach (Roberts 1998a, 146-8;Navia 2009) consolidating neoliberal economic policies rather than questioning them.This reflected a structural transformation of the party coalition.In the 1990s, its leftist arm, the Socialist party, turned from a party of 'mass mobilization' to a 'party of notables with only the minimal development of grass roots organizations.Most importantly, mass urban and rural mobilizations of the mid 1980s has been replaced by mass-mediated 'pursuit of individual voters with diverse and fluid political identities' and 'the dilution of ideology' (Roberts 1998a).Through the mass-media logic, they have advanced policies that favour equality of opportunities (job training, house subsidies loans, institutional conditions) rather than equality of conditions, and fed their voters with the services provided by the modernized criminal courts system (Dammert 2006;CEJA 2008b), the police (Ramos and Guzman de Luigi 2000) and focalized welfare (Raczyncki 1999).
On the coercive front, the government and party experts have decided to reinforce the police, the courts and the prisons, increasing their budgets and personnel but also converting the inefficient providers of penal services.The national police forces turned from national security watchdogs to public providers of security services, to citizens-consumers (Dammert 2006;Frühling 2009).Within the criminal justice system, Concertación administrations led due process and efficiency reforms in the criminal procedure (Langer 2007).Prisons were privatized following the experts of the Lagos' administration.The same managerial rationalization was instituted in the welfare ministries and agencies, even if it presented lesser overall coherence.Here the ideologically and organizationally renovated centre-left-Coalition continued a major use of coercion -no longer by the military, but by an expanded and legitimated criminal justice.On the welfare front it remained less generous.
But besides political party practices in democracy, the initial penal welfare arrangements were also reinforced by the highly conservative character of urban grassroots mobilizations.Chile experienced important urban poor mobilizations after the transition to the neoliberal regime.However, they have been of a nature that reinforced the original neoliberal designs.In the Chilean case, the historically combative urban proletariat and subproletariat of Santiago, repressed during the first decade of the dictatorship, came back in 1983, being the core of the protests and the most radicalized sector.Many were recent immigrants displaced from the previously socialized countryside, and were now demanding housing and land in the city.Centre-left political parties capitalized on those struggles to produce democratic openings (Garreton 1998).After the return of democracy, Santiago experienced a steady stream of urban protests.However by this time, 'the homeless of the 60s ('los sin casa') have been replaced by families of propertied poor as the principal subjects of urban popular movements' (Sabattini and Wormald 2003, 59).Their demands now pointed to 'quality of life issues, to urban equipment and infrastructure, public transport, protection against crime, and against public works that threaten the quality of life of their neighborhood, affect their health, and their main economic investment, their house' (Sabbattini and Wormald 2003, 59) that are beyond individual or family means (De la Maza 1999, 393).The protests mobilized 'territorial identities' (Sabattini and Wormald,73), abandoned the communist party -seen as too radical and ineffective -and benefited from the 'urban development,' housing, and 'security policies' promoted nationally by Concertación and by the right-wing parties in Santiago (Barozet 2003) during the 1990s.Falling into what Dosh calls the 'security trap of social movements' (Dosh 2009), they not only reject general and more generous welfare policies that may benefit the newcomers to the city, but they also favour police protection of their neighbourhood and penal harassment of the zones of recent immigration; in the south quarters, such as La Pintana, the incarceration rate almost triplicates the national average (610/100,000).In Chile, urban poor mobilizations consolidated the neoliberal policies of subsidiary targeted welfare and expanded penal surveillance over the marginal populations.The recent mobilization of working and middle-class youths asking for public education does not impact on these policies (Luna 2011).Things have been very different in Argentina, at the beginning and later on.

Argentina: democratic neoliberalization, undermined by neo-populist parties and grassroots mobilization
In the democratically imposed neoliberalism of Argentina, we observe participation of labour in the structural adjustment coalition, which produced, paradoxically, a reduced welfare and weak role of state coercion.Organized labour and private local capital participated in the reformist coalition that implemented the structural adjustment led by Menem (1990).Given the backing of unions, this structural adjustment faced less organized resistance than in Chile (and Peru).While organized labour blocked labour law reforms, it accepted a drastic reduction in welfare, leading to an exclusion of informal workers and urban marginal groups from welfare provision (Etchemendy 2004).Thus, initially poor relief was almost inexistent, coexisting with the generalization of private strategies of survival, in microenterprises and personal services companies where severance payments were invested.
On the penal/repressive front, the inexistence of a leftist radical opposition (disarmed a decade earlier during the 1976-1983 military dictatorship) led to a very small role for the military.The police, courts and prison had an important but ultimately secondary role in enforcing the structural adjustment programmes, being in no way as active as they have been in Chile (or in Peru as we will see later).These penal bureaucracies remained untouched in terms of resources and personnel, except for a managerial rationalization and mild reform used by the central government to expand its control over them.
Two economic and urban effects of structural adjustments will be consequential for the evolution of these initial welfare and penal policies as affected by party organizations and urban mobilization.At the economic level, the protection of the local industrial capitalists, concentrated in Buenos Aires, led to reduced reinvestment and progressive loss of competitiveness (Etchemendy 2005), and to a much less diversified economy.This, combined with the privatization of public enterprises, and the shrinking of public employment, brought a generalized impoverishment of the traditional working and middle classes, and greater poverty for the urban informal sector -traditional providers of work for small industries and of cheap services to the also declining middle classes (Portes and Hoffman 2003).In 2003 Buenos Aires had 60 per cent of inhabitants under the line of poverty (Cerruti and Grimson 2004).The abandonment of informal workers and the urban poor left a marginalized urban population ripe for political mobilization, first by clientelistic networks, and later on by grassroots organizations.
In this case, the historically centre-left Partido Justicialista (PJ), after implementing the structural adjustments in the economy, became a neopopulist party; it followed neoliberal principles but secured followers through massive clientelistic practices.Clientelism and internal divisions with the PJ during the 1990s produced an initial expansion of welfare, but also an intensification of coercion over the urban poor, increasing penal severity and surveillance and criminalizing protests.The struggle since 1995 within the PJ, in particular between Buenos Aires Province governor E. Duhalde and President Menem, eroded the early minimal welfare scheme.Duhalde developed an extensive clientelistic network during the 1990s, organizing a parallel welfare bureaucracy (Acuña, Kessler et al. 2002).This was the first step within the general mutation of Peronism to a machine politics party based on an ever expanding 'clientelistic' welfare (Levitsky 2005).The new welfare-clientelism networks, however, did not lead to universalist welfare, not even under the Alliance between the Unión Cívica Radical and Front for a Country in Solidarity (FREPASO), who questioned not 'the economic model, but the lack of transparency and corruption' of the Menemist administration (Svampa 2005, 55).A more universalistic policy emerged only in post-2003, as a result of demands of urban organized civil society.
In the meantime, clientelistic Peronism gave new uses to the penal branches of the state.First, following a typical strategy of penal populism, in the middle of Menem's second term, the government in Buenos Aires and the federal government passed more repressive laws for common crimes (Sozzo 2002).Second, the neoliberal Peronist Party instrumentalized the historically politicized police forces (Andersen 2002), participating in the gains of police corruption networks, investing them in internal party struggles (Sain 2008).Finally, the Peronist bosses, in particular those controlling the Buenos Aires suburbia, deployed the police to keep the destitute in place (Ranguni 2010) and repress the Piqueteros -the re-organized unemployed and the urban poor excluded from the periphery of major cities and economic enclaves (Svampa 2005).
The combination of structural adjustment poverty effects, the reduced welfare and the harshening of police and the criminalization of poverty is preparing the ground for a counter-hegemonic mobilization by the urban poor that will modify the policies even more.In Argentina grassroots and working class organization will reverse the original targeted welfare, and will systematically question the criminalization of protest and human rights violations of the urban poor (Tiscornia 2007).The very limited welfare that expanded during the 1990s left the urban poor without protection and abandoned informal workers and impoverished middle sectors.This was com-bined with a severe economic retraction and a general deterioration of the labour market in terms of informalization.After the implosion of the urban economy, and minimal subsistence assistance, the urban poor started demanding jobs or income from the state.By reviving their organizational structures, developed in the 1980s during the occupation of lands, and their demands for housing, and resorting to the old union struggle and neighbourhood identities, not destroyed as in Chile and Peru, the Piqueteros have increased their capacity to mobilize and make new demands, reversing the welfare scenario in Argentina.As 'an outcome of protest and organizing, social policy for labour market outsiders [unemployed and informal sectors] underwent crucial changes, and expanded massively, … [while] in kind benefits have lost centrality vis-à-vis more generous cash transfers ... favouring [also] access to health and pensions' (Garay 2007, 302).Even if some piqueteros organizations joined the new governing coalition led by leftist P.J. Nestor Kirchner (Garay 2007, 315), other sectors have sustained such transformative mobilization (Svampa 2011).
The result of these anti-systemic mobilization is an expanded but still 'segmented' welfare, where clientelistic networks, civil society organizations and municipalities have a share in providing assistance.Here, in contrast to the Chilean centralized workfare system, welfare provision through these different channels is deployed not to discipline the atomized working class -the welfare discipline Wacquant mentions -but to accrue political capital, to produce votes through party networks or to gain mobilization capacity for grassroots organizations.However, as grassroots contestation grew it has called forth a renewed expansion of repression and police harassment, directed or tolerated by the PJ.This expanded repression feeds the activism for the respect for the life and liberty of the urban poor who end up in the hands of the police, courts or prisons (i.e.Tiscornia 2008).This leads to a constant questioning and delegitimation of the criminalization of urban poverty and of the repression of protests.

Peru: semi-authoritarian repression and small welfare continued by clientelism and weak mobilization
In Peru, neoliberalism was imposed by a semi-authoritarian regime, thus presenting features of these two policy regimes: high use of coercion and coercive bureaucracies -like in authoritarian Chile -and a highly reduced welfare -like in democratic Argentina.These initial features will remain in place thanks to clientelist populism and an urban grassroots mobilization that does not question those policies.Here Fujimori -a political outsiderimplemented neoliberalism within a context of radicalized leftist mobiliza-tion by Shining Path and after a severe economic crisis that atomized urban civil society.After two years in power, and once he had put in place major structural transformations, in alliance with the docile sectors of Army and the intelligence service, Fujimori performed a self-coup, suspending the constitution, dissolving congress, and targeting the opposition' (Beggar 2005).After the coup, he continued attacking organized labour, the traditional constituencies of APRA.As in authoritarian Chile, Fujimori 'ignored union-backed workers yet deliberately attempted to construct a political constituency among unorganized workers in the informal sector and urban poor' (Etchemendy 2004b, 286).But, unlike Chile, here Fujimori's neoliberal advisers were unable to institutionalize their workfare recipes in a national bureaucratic organization (Roberts 2005).By contrast, clientelism, like in Argentina, displaced bureaucratic impersonal allocation of welfare.Fujimori 'controlled programs of poverty relief, food assistance, school construction and milk supplements for children and pregnant women, […] transforming them into instruments of state patronage, targeting the atomized urban poor (Castellanos and Pereyra 2004).The semi-democratic nature of the regime explains such initial reduced but clientelistic welfare.On the repressive side, the presence of Shining Path legitimated the expansion of the military over the country and urban zones, allowing Fujimori to repress the working classes and urban and rural poor (Beggar 2005) and to neutralize opponents through sheer fear (Burt 2006).
The neoliberal restructuring here brought important economic and urban changes that will interact with party transformations and condition the reactions of the urban poor to these policies.De-industrialization, factory moves, and privatization increased unemployment and the more flexible work regime led to increased poverty.In relation to the urban system, Lima received a great number of rural immigrants escaping guerrilla and military repression, or sold their lands for lack of competitiveness in the market.The recently urbanized, atomized and ethnically diverse urban migrants started to compete with the already established urban poor and recently impoverished urban workers, who had been owners of their houses since the Velazco Alvarado governments (1972)(1973)(1974)(1975)(1976)(1977)(1978).The old marginal urban dwellers, in turn, lost much of their organizational structures and local leadership, decimated by the economic crisis and by Shining Path and military interventions in the late 1980s and early 1990s.Within Peru's personalistic but unified parties, Fujimori's successors, presidents Toledo (2000Toledo ( -2005) ) and Garcia (2006)(2007)(2008)(2009)(2010)(2011), preserved the focalized but scarce welfare.Despite its brief opening to a more rationalized and expanded welfare, Toledo continued the neo-populist line of Fujimori, implementing a highly selective but very humble increase in health, food assistance, housing and urban renewal for the urban poor (Barr 2003(Barr , 1165)).After him, Alan Garcia continued with the same flexibilized labour, and the promotion of micro-entrepreneurialism, and targeted subsistence food programmes (Copestake 2008), but did not change routes.Here highly concentrated control of clientelistic networks by the executive branch, instead of producing an increase in welfare, as occurred in Argentina given the divisions in the Peronist party, prevented its expansion.
The expanded and highly repressive penal sector also remained in place.The National Police, in the meantime, converted themselves into specialists in the war against drug trafficking, targeting growers in rural areas and dealers in urban ones, deploying their own 'security plan' based on greater surveillance of problematic neighbourhoods, increasing both personnel and resources.Here, as in Chile, the post-authoritarian administration attempted to reform the police (Basombrío Iglesias 2004), but given the high levels of corruption and the protection by political figures (Costa and Neild 2007), reform was very limited (Basombrío Iglesias 2006).Attempts to reform criminal courts toward greater efficiency failed, while prisons filled with prisoners waiting for trial (Ponton and Durán 2006).The institutional instability, the enormous police and military corporative power as well as corruption, prevented reforms but also pushed the urban poor to continue with neighbourhood patrols (serenazgos) to provide security for themselves.The military and the police are still the core of Humala's security policies (Toche 2011).
In these conditions of extremely limited welfare and highly violent but inefficient penal state, urban grassroots organizations engaged in protests and mobilization that consolidated the initial limited welfare and fragmented security system.As in Chile, the urban poor who organized themselves are mainly property owners that engaged in very private strategies of social reproduction, based on the preservation or improvement of their individual households (Dosh 2007) and in local protection from crime (Pereyra 2003).
Here, the weakened links between the majority of the poor urban residents and political parties and the general retrenchment of the state led to very important privatization of aid and protection.The most common demand has been, like in Chile, security, and the greater urban mobilizations have been oriented to reduce and prevent crime, but in this case, not by collaborating with the police, but through citizens policing themselves (Pereyra 2003).The more organized neighbourhoods defend themselves from corrupt police forces and from crime that comes, according to many residents, from the neighbourhoods of recent immigrants from the countryside.The distant presence of the state and the selective intervention of clientelism breaks possible horizontal links, leading to an atomized, fractured and privatized marginal urban civil society whose demands regarding urban renewal and security consolidates the initial penal-welfare policies in place since Fujimori.Outside Lima, protests against privatization of public companies and foreign direct investments have questioned neoliberal economic recipes (Arce 2008).

Concluding remarks: understanding the particular within the general trends of neoliberalism
In this piece I have tried to advance our understanding of the variations in the penal and welfare neoliberal state in Latin America and on how neoliberal penal and welfare policies are shaped by local political processes that differ enormously from those in the global north and that give specific declinations to different neoliberal penal and welfare states in Latin America.The comparative approach showed the causal relevance of the regime that governs the transition, party ideas, and most important, party organization and the target and strength of urban poor mobilization.By observing how they interacted we can better understand the specific shapes and extent of the dismantling and downsizing of the 'the Left Hand' of the state (welfare) and the expansion of the state's '"Right Hand", the police, justice, correctional administration' (Wacquant 2009, 6).This work coincides with other studies that depict penal policies as central within the neoliberal governance of the urban poor in the region (Godoy 2005; Muller 2011).At the same time it serves us to qualify some general characterizations of the neoliberal penal state in Latin America.Muller (2012) argues that the neoliberal penal state in Latin America differs from that in the global north not only in its harsher effects -propelled by greater inequality, police violence and racism -but also in that it involves highly informal bureaucracies and 'incorporates a variety of actors operating beyond the formal bureaucratic field -vigilante groups, death squads, militias -into its workings of power' (Muller 2011, 3).The cases referred to here show that bureaucratic informality and proliferation of agents of coercion are absent in Chile, and that bureaucratic informality and pluralization of agents of coercion do not necessarily come together.While in Argentina we have highly informal and weak bureaucracies, there is no proliferation of private agents of coercion as in Peru.Moreover, I would argue that the analysis of types of transition, political party ideas and practices and types of urban poor mobilization studies are relevant to understanding those differences in contemporary bureaucratic informality and the proliferation of coercive and surveillance agents in neoliberal times discussed by Muller.In closing I must recognize that here I have dealt with issues of general intensity, extent and general orientation in welfare and penal policies.The precise contents of the penal and effects and welfare policies in each country is beyond the reach of this paper.However, even at this general level, the political dimensions analysed are systematically connected to differences between neoliberal penal and welfare leviathans, even within the contours defined by the economically conditioned and politically weakened democracies of neoliberalism (Weyland 2008).
<hathazy@berkeley.edu> is a PhD candidate in the Department of Sociology at the University of California Berkeley.His main interests are the state, law and experts, deploying qualitative methods, and privileging historical comparative approaches.His recent publications include (2012) 'Enchanting Bureaucracy: Symbolic Violence and the (Re)production of Charismatic Authority in a Police Apparatus', in International Sociology, 27(6); and (2013) 'Fighting for a Democratic Police: Politics, Experts and Bureaucrats in the Transformation of the Police in Postauthoritarian Argentina and Chile', in Comparative Sociology, 12(4).

Table 1 .
Variables, conditions and initial and ulterior penal and welfare policies in Chile,