Easing the tension between the state and the market? Developing social protection and labour law during Latin American industrialization

Abstract The onset of industrialization across Latin America put the social question squarely on the agenda of policy-makers. Although Latin American countries in many respects did not meet the conditions of socio-economic and political development that proved to be the prerequisite for welfare state creation in Western Europe, the early Latin American industrializers decided to broadly follow the style and function of the European model introducing social rights and a rather truncated notion of citizenship. Nevertheless, those early policy decisions have proved to lastingly impact social, economic and political conditions across the region with central aspects of social protection and labour law having limited reach due to deficiencies in state capacity, the rule of law and high levels of labour informality. Discourses on the development of citizenship, social protection and labour law are inextricably linked to the complex processes that make up industrialization entailing thorough economic and political transformations and struggles within society. The resultant so-called social question has ever since the inception of industrialization given rise to intense debates on social inclusion, welfare state creation and the design thereof. This paper focuses on state-society relations during the early stages of industrialization in Latin America until the 1980s, predominantly analysing the cases of Argentina, Brazil, Chile and Colombia.

ABOUT THE AUTHOR Ulf Thoene, PhD, is an associate professor at the Universidad de La Sabana, Bogotá, Colombia.He completed his doctoral studies at the University of Warwick in 2013.His research interests focus on Business Ethics, Social Policy, and Industrial Relations in Latin America with a particular focus on Colombia.Prior to becoming an academic Ulf served in the German Armed Forces and worked in the private sector.

PUBLIC INTEREST STATEMENT
Discourses on the development of citizenship, social protection and labour law are inextricably linked to the complex processes that make up industrialization entailing thorough economic and political transformations and struggles within society.The resultant so-called social question has ever since the inception of industrialization given rise to intense debates on social inclusion, welfare state creation and the design thereof.This paper focuses on state-society relations during the early stages of industrialization in Latin America until the 1980s, predominantly analysing the cases of Argentina, Brazil, Chile and Colombia.

Introduction
The onset of industrialization across Latin America put the social question squarely on the agenda of policy-makers.Although Latin American countries in many respects did not meet the conditions of socio-economic and political development that proved to be the prerequisite for welfare state creation in Western Europe, the early Latin American industrializers decided to broadly follow the style and function of the European model introducing social rights and a rather truncated notion of citizenship.Nevertheless, those early policy decisions have proved to lastingly impact social, economic and political conditions across the region with central aspects of social protection and labour law having limited reach due to deficiencies in state capacity, the rule of law and high levels of labour informality (Thoene, 2015).
The central and most appealing quality "of industrialization [lies] in sustaining high rates of economic growth leading to high levels of income per capita" (Álvarez & Brando, 2019, 2, italics in original).Analysing aspects of industrialization and their relationship to social protection, labour law and the formation of citizenship over time in the case of Latin America is of particular interest due to the fact that despite the transformative power of industrialization on overall society, the impact on politics was relatively muted.One would expect that the economic transition away from agriculture and the dependence on commodity exports towards industrialization unleashed forces impinging on the fabric of society and setting in train structural change.
Following this introductory passage, this paper traces the development of social protection and labour law in the region, highlighting the evolutionary steps of the welfare protective regime as well as the political bargains and power struggles during the broad historical period that started with the onset of industrialization and ended with the 1980s Latin American debt crisis.This historical period was marked by the political necessity of coming to terms with the social question and managing incipient industrialization.To a large extent, these historical phases explain the trajectory that the welfare state has taken in Latin America, and that has proved to be stable into the contemporary era as evidenced by today's prevailing design structure of social protection and labour law.Empirical data predominantly stems from Argentinean, Brazilian, Chilean and Colombian sources.

Historical and institutional foundations
Looking beyond the surface of the political realities of the Latin American polity in the postindependence environment and going forward, we learn that "South American electoral traditions formed before consolidating practices and understandings of civic and political citizenship" (Adelman, 2003, p. 283).That lack of consolidation led to social and political instability due to a perceived sense of injustice among large parts of the population.Furthermore, Latin America adopted a capitalist market-based economic system "[applying] free-market dogmas without inscribing an underlying juridical order capable of legitimising public authority" (Adelman & Centeno, 2002, p. 141).
Against this background, it is important to point out that two years before the introduction of the 1919 Weimar Constitution in Germany, the Mexican Constitution of 1917 had elevated social rights to the level of individual, constitutional rights.The case of Mexico serves as an example for the "recognition of the so-called social problem [that] had found expression throughout Latin America as early as the 1890s" (Halperín-Donghi, 2010, p. 29).Halperín-Donghi further argues that in the early years of the 20th century the question of social inequality had found its way into the political arena rather than being seen as solely "[leading] to violent disturbances of the social and political order" (2010, p. 29).In fact, constitutional developments in Mexico subsequently played a key role for the institutionalized proliferation of legal commitments to worker protection in a host of other Latin American countries (Bensusán, 2007, p. 7).
Moreover, the examples of the relatively early and well-intentioned adoption of the social rights issue into political and legal discourse on the one hand, and the lack of real and progressive social policy outcomes on the other hand, serve as stark reminders of gaps in implementation and consolidation.Flaws in the development of the legal system, weak democratic political institutions and inadequate welfare systems were stumbling blocks on the road towards realizing constitutional aspirations.
Nevertheless, during the early years of the 20th century some activist Argentinean judges acted as the first branch of government to formulate answers to the social question, as they accepted cases of workers taking their employers to court over occupational accidents (Schjolden, 2009, p. 91).In the absence of legislative action and the passivity of the executive, such instances of "court-made labour law" began to advance the notion of citizenship and more specifically of social rights before the passing of legislation by Congress in 1915 (Schjolden, 2009, p. 98).In other words, the judges faced a new social and legal environment in which plaintiffs could only use the Argentine Civil Code to press for their rights claims because no specific legal mechanism to address the consequences of industrialization for individual workers had been developed (Schjolden, 2009, p. 99).
Furthermore, "the long, dark shadow cast by predatory colonial elites" (Cypher, 2018, p. 25) and the legacy of Spanish and Portuguese colonialism "[leaving] behind immature capitalist classes" (Rapley, 2007, p. 29) had imposed upon post-independent Latin America an institutional framework that paid little heed to the fundamental civil right of private property, thereby discouraging capital investment (Acemoglu, 2003, p. 629).Moreover, "little investment was made in the human capital of the majority of the population and access to key resources was concentrated in the hands of a few", which severely limited participation in politics and by extension in law making, too (Acemoglu, 2003, p. 629).
Thus, the playing field was far from level in the economic sphere, and the socio-political and legal sphere lacked the rule of law.The economic and political institutions governing society were extractive rather than inclusive (Acemoglu & Robinson, 2012).With respect to the deficiencies in consolidation and legitimacy of the legal and political order, Bendix argues that "[f]ormal equality before the law at first benefits only those whose social and economic independence enables them to take advantage of their legal rights" (1964, p. 123).On a theoretical level, one could expect that the Latin American republics hence sought to empower the citizenry with a degree of social and economic autonomy, in order to subsequently extend the same powers to wider sections of society.However, the mass of empirical evidence on which Adelman and Centeno (2002) and Acemoglu (2003) base their arguments points in another direction.
Bendix ' (1964) argument rather fits the circumstances prevailing during the nascent stages of the republics.In order to approximate an ideal such as equality before the law, a body of rules and regulations had to be created, enabling increasing numbers of individuals to incorporate into the growing society, provided they subscribed to rule-based order.However, due to the social and political legacy of the colonial past, the weakness of democratic governance and the fluctuating nature of Latin American integration into the global economy, the "rights attached to citizenship have […] often been of little practical value to the inhabitants of the region" (Roberts, 1996, p. 38).In the Latin American case, the forces of path dependency and political inertia feature and persist in a particularly open and stable way.
Moreover, suffrage tended to be limited to men who could comply with certain voting rights regulations based on their income and the taxes they paid.Larger parts of the population received the right to vote in various phases over the following decades, with women's suffrage predominantly being introduced as late as the mid-twentieth century.Commencing in the 1930s, the expansion of the franchise resulted in the popular sectors advancing claims to "broader civic and political liberties, as well as social rights" (Adelman, 2003, p. 290).A brief focus on the issue of universal suffrage is important because once that vital right had been achieved, popular pressure sought to extend the notion of social citizenship to "full participation in public life [and] to create a legitimate democratic form of government" (Adelman, 2003, p. 290-291).
Nevertheless, this chapter focuses on the period beginning with the first indications of the social question in the wake of early developments in urbanization and in the industrial economy in the Latin American region.The changes within society during that period challenged the established political order, opened pathways towards the reconfiguration of socio-economic development, and "unleashed forces that needed to be tamed" (Fukuyama, 2012).

Industrialization and urbanization
Focusing first on changes in the economic landscape over time, one can argue that with economic growth taking hold and processes of urbanization speeding up, "wage employment and […] dependence on the market for survival have brought to the fore key elements in contemporary citizenship such as labour rights and social security" (Roberts, 1996, p. 44).Against the background of twentieth century history, we can divide Latin American countries into three groups regarding their rates of industrialization and urbanization.
In a nutshell, Roberts categorizes Argentina, Chile and Uruguay into a group of countries that had reached high levels of income by the 1940s (1996, p. 44-46).The countries also first made advances in terms of citizenship and welfare institutions.The second group includes the currently three most populous Latin American countries Brazil, Colombia and Mexico.Roberts claims that by the 1980s those countries had eventually caught up with the first group in terms of income, and half their populations lived in cities (1996, p. 45).A third group labelled "slow developers" comprises such countries as Bolivia, El Salvador, Honduras and Guatemala (Roberts, 1996, p. 46).Those countries "were among the least urbanized and economically developed", and their economic specializations in agriculture and mining had not changed much over the period covered (Roberts, 1996, p. 46).
Urbanization is an important analytical category because over time most Latin American countries have become highly urbanized.What is more, waves of urbanization brought with them individuals who had not experienced long-term wage employment in the countryside and who tended to pursue a range of different occupational activities in the cities because of a lack of stable employment in the industrial sectors.The activities of the new urban classes ranged from wage employment, to occasional day work and self-employment.It is interesting to note that the only migrant labourers who had had prior experience of organized labour were miners and plantation workers (Roberts, 1996, p. 47).
Urbanization had an immediate effect on citizenship because the new and growing industrial complexes in the cities "gave a certain bargaining power to workers in the key sectors of the economy" and enabled the establishment of trade unions (Roberts, 1996, p. 50).However, whilst one might intuitively expect that such forms of non-state political organization would strengthen citizenship, the large urban informal sectors in effect limited the size, hence the political bargaining power, which the unions could bring to the table (Roberts, 1996, p. 51).Furthermore, whilst in such countries as Argentina urbanization was rapid, other parts of Latin America progressed slowly in rural-urban migration, which complicated the formation of a sizeable and thus potentially significant urban working class.For instance, during the early-1920s only six out of a hundred Colombians populated "the three largest cities" (Henderson, 2001, p. 118).Yet following dynamic urbanization during the 20th century, three quarters of the population was urban, having left the rural zones of the country since 1980 (Henderson, 2001, 118, p. 326).Fundamentally, the process of urbanization as the "mother of transition" in demographic, economic and cultural terms was completed between 1940and 1990(Rueda-Plata, 1999).
Moreover, the development of capitalism and industrialization took a different path in Latin America than in Europe.According to Centeno and López-Alves, the "timing of industrialization and the absence of geopolitical competition produced a very different form of industrial capitalism than Bismarckian, Meiji, or Stalinist 'catch-ups'" (2001, p. 8).According to Gootenberg, whose claims could be understood through lens of dependency theory, which "argued that the first world actually hindered the emergence from poverty of the third world" (Rapley, 2007, p. 26), Latin American countries did not follow "[Germany's] strong state-heavy industrial path but were precluded by free-trade imperialism, their entrenched 'comprador' elites [and] neocolonial social structures" (2001, p. 67).Nevertheless, countries such as Argentina and Uruguay stood out as examples of "early developers", due to stellar economic growth towards the late 19 th and the early 20 th centuries, as well as reaching very high levels of urbanization over following decades (Draibe & Riesco, 2007, 81, italics in original;26).

Developing social protection and labour law
Early industrialization during the first half of the twentieth century led Latin American polities to pass legislation in order to protect life risks, to regulate the employment relationship and to define labour standards together with the early adoption of social security schemes in the 1920s.That process sped up during the "1940s and 1950s linked to Import Substitution Industrialization [ISI]" (Huber, Mustillo, & Stephens, 2008, p. 423).Rodríguez, for example, traces the institutional inception of Colombian labour law that addressed such labour and social security aspects as the resolution of health and safety concerns, the regulation of strike action, occupational rest periods, and obligatory insurance for workers to between 1915 and 1926 (2001, p. 615).A further legislative push in Colombia lasting until 1957 introduced an autonomous branch of judicial power for labour cases, regulation regarding the funding of unions, and established the organizational structure under which labour courts still operate today (Rodríguez, 2001, p. 616-617).
Thus, by the 1960s most of the Latin American region had established protective schemes with varying degrees of coverage and administrative capacities (Mesa-Lago, 1989, p. 6), which was a development that must also be understood against the background of the emergence and increasing popularity of modernization theory, which "sought to identify the conditions that had given rise to development in the first world, and sought to specify where and why these were lacking in the third world" being favoured mostly by "Westernizing elites" (Rapley, 2007, 24, p. 26).However, access to welfare benefits was always tied to a formal employment relationship, and the first groups to benefit from social and labour protection were individuals employed in the public sector, military personnel and workers in the heavily export-oriented mining industries.Seen through the conceptual lens of the private theory of regulation, such state capture by certain groups in society "[shifts] resources from the private to the public sector [enlarging] the size of the administrative component of government", and disadvantaging groups that unsuccessfully pressed for inclusion in the protective regime, but who were nonetheless compelled to share the financial burden either via taxation or by missing out as ineligible recipients of state spending (Wilson, 1975, p. 103).
Whilst relevant for the early phase of industrialization, decades later Portes and Hoffman argue that, as opposed to advanced industrialized societies, such social classes as the Latin American "'proletariat' […] are segmented by their limited incorporation into the fully monetized, legally regulated economy" (2003, p. 44).That claim stresses the Latin American labour movement's historical weakness in terms of broadening the coverage of public social policies, so that the notion of state capture applied above must be modified.In other words, only few societal groups were actually successful in seeking access to early manifestations of the welfare protective regime.Some occupational groups such as in the Chilean manufacturing and mining industries and in the public sector earned higher salaries than agricultural workers (Huber, 1988, p. 23).Due to the geographical distance between the Chilean mining areas and the country's cities, the organizational strength of mine workers remained by and large a local phenomenon and did not enable urban workers to reap comparable social and political benefits (Huber, 1988, p. 23).The Chilean labour code, which was established in 1931, aimed to "fragment the labour movement" and to keep the unions out of political decision-making (Couso, Lovera-Parmo, Guiloff, & Coddou, 2011, 212;Faúndez, 2011, 115, my translation).In the process, the labour movement was relegated to the role of "a spectator rather than an active participant", whereas some of the main industry federations acted as "voting members of several semi-autonomous key financial policy institutions" (Angell, 1972, 41, 79, 81).That latter observation already hints at the fact that "many pieces of social legislation were indeed enacted in the context of electoral tactics or promises" rather than being motivated by a political desire to strike sustainable and equitable social bargains (Drèze & Sen, 2002, p. 230).Whereas in Chile the state used labour legislation to shape state-labour relations in its pursuit of establishing the control and "incorporation of the […] labour movement", successive Peruvian executive branches of government sought incorporation via a political party, because, until the 1970s at least, "control over unions [was] not as extensively embodied in legal statute […] to the same extent as in many other Latin American countries" (Cook, 2007, 106;Angell, 1980, p. 10).
Similarly, a 1955 law "instituted a rigorous process of [electoral] re-registration" effectively excluding one half of Brazilian citizens of voting age from participating in elections (Holston, 2008, p. 103-104).At mid-century almost two out of every three Brazilian citizens were living in rural areas where illiteracy rates were high and where the developments of labour legislation pioneered in the cities were generally ignored (Holston, 2008, p. 105).Consequently, "rural workers […] began to abandon the countryside for the city in unprecedented numbers" from the 1950s onwards in order to seek the prospects of economic stability offered by higher paid jobs (Holston, 2008, p. 104).As an unintended consequence of internal rural-urban migration, "manipulation of the electorate became more difficult" (Skidmore, 1967, p. 3).
The political and economic elites did not only pass labour legislation aimed at the fragmentation of the labour movement, but they also hardly concealed the fact that different societal groups enjoyed access to varying degrees of political rights.Thus, the majority of Brazilian citizens merely enjoyed a truncated set of rights rather than the complete canon of civil, political and social rights that inject meaning into the notion of citizenship.The uneven distribution of rights claims also directly affected an individual's opportunities in the labour market, because the empowering and protective elements that resulted from the evolution of citizenship were evidently limited to a relatively small section of the population.
Put differently, the complementarity of different rights and social policies crystallizes into a key developmental theme in as much as there exist obvious links between the spread of literacy, the exercise of rights, and trends towards urbanization, for instance, which over time resulted in entrenched patterns of socio-economic hardship rather than in steps towards the alleviation of social strains.Furthermore, the Chilean labour code detailed the contents of employment contracts and "determined the right to a minimum vacation period, the redundancy procedure, family allowances and the minimum wage" albeit with an inflation-calming component (Faúndez, 2011, 88-89, my translation).Nevertheless, despite inscribing into law a range of social rights for certain occupational groups, the labour code-imposed restrictions on collective bargaining and strike action (Faúndez, 2011, p. 88), which signified a curtailment of political activism, civil rights, and hence a limitation of citizenship.Similar developments took place in Colombia where the year 1922 saw the introduction of social security legislation and 1924 witnessed the creation of an Office of Labour (Henderson, 2001, p. 160).As a case in point, two years later "it became law that workers could not be required to work on Sundays" (Henderson, 2001, p. 160).
Nevertheless, analogous to the Chilean case, "what the [Colombian] government seemed to offer with one hand it more than took back with the other" (Henderson, 2001, p. 160).Emphasizing that point was the passivity of the Colombian government in the face of foreign companies' labour law breaches in the oil and banana industries throughout the 1920s (Henderson, 2001, 160, p. 165).It took the Colombian government until the mid-1940s to push through further labour reform and policies of "social welfare through the guiding hand of the state" (Henderson, 2001, p. 283).
Although workers were now given "the right to written contracts, severance and retirement pay, sickness and accident insurance", the possibilities for strike action and collective bargaining between workers and businesses in the absence of state interference were further curtailed, as the government feared the emergence of an influential party on the far left of the political spectrum and thus intended to exclude unions from the political process (Henderson, 2001, 283, p. 336).
Equally, in spite of the elite's charges against organized labour, and despite the fact that unions would not appear as "a considerable voice in public affairs" until the mid-1960s, workers actually "enjoyed better physical conditions during the 1950s than at any previous time in Colombian history" (Henderson, 2001, 337, p. 406).This presents us with a somewhat paradoxical situation, as the military-led government of Gustavo Rojas Pinilla (1953)(1954)(1955)(1956)(1957), presided over concrete advances in terms of social rights as well as the introduction of universal suffrage in 1957 (López-Ocampo, 2012).However, the triad of political, civil and social rights was only superficially developed, and it was held back by severe socio-economic inequality and the Colombian elites' refusal to face up to the politically explosive question of land ownership and distribution (Gómez-García, 2009, p. 141).
We gain additional insights into the discontinuities of political governance and how such ruptures affect the quality of citizenship.Even though Argentineans had achieved universal suffrage in 1947, the country enjoyed a severely limited form of citizenship during military rule in the 1970s and 1980s.Although the dictatorship left largely untouched certain social rights, for parts of the populace political rights were profoundly violated.That was particularly the case in the areas of freedom of expression and association with respect to demonstrations and union marches.Under those circumstances, a significant proportion of citizens tended to favour such social rights as schooling or basic health care over political rights.Hence, the evolution of citizenship in Latin America has been fragile and non-linear, mirroring the instability of several spheres of social life.Political upheavals and phases of economic boom and bust have been inextricably linked to the erratic convergence and frequent dislocation of different strands of citizenship rights over time.
Furthermore, in late 1949 a bloody civil war-La Violencia-broke out in Colombia.Henderson claims that: "most citizens managed […] to avoid the politically inspired Violencia" as they increasingly tended to live in urban areas rather than the more violent-prone countryside (2001, 326, p. 341).Henderson further argues that "most Colombian institutions grew significantly stronger during the years of Violencia […] helping insulate Colombia from the sort of populist excesses that troubled the postwar economies and politics of Argentina, Peru, and other Latin American nations" (2001, 327, p. 339).The 1948 assassination of the popular politician and orator Jorge Eliécer Gaitán (1903Gaitán ( -1948))-the "potential Perón of Colombia"-who drew strong support from the middle and lower classes who expected him "to pursue populist and redistributive policies" (Acemoglu & Robinson, 2006, p. 52), and subsequent heavy rioting in the streets of Bogotá-known as the Bogotazo-"demonstrated the revolutionary potential of the masses" and compelled both government and opposition into turning their attention to "the needs of the nation's poorest citizens" (Henderson, 2001, p. 340).
Actually, it can be argued that Gaitán's death foreshadowed the murder of Luis Carlos Galán  in 1989, another prominent hopeful of the Colombian Liberal Party (Acemoglu & Robinson, 2006, p. 52).The direct aftermath of the 1948 Bogotazo saw the inception of the Servicio Nacional de Aprendizaje (SENA)-National Training Service-, which has over the decades evolved into a key Colombian labour market institution and a provider of occupational training, as well as an example of government orchestrated cooperation between the state, businesses and schoolleavers or workers.
Furthermore, in 1949 the Colombian executive determined a new round of social legislation in the areas of social insurance and labour law "mandating severance pay, housing subsidies, paid vacations, and private recreational clubs for employees" (Henderson, 2001, p. 341).It is crucial to highlight those developments as their inception has had a lasting effect on social stratification, the welfare of Colombian citizens, and current political debates on labour market institutions, as well as the social protection system.
Many of the policies flowing out of the 1949 legislative achievements have been funded mainly by payroll taxes levied on holders of formal jobs.Unsurprisingly, the issue of labour costs has become subject to heated debate with respect to regulatory approaches to the informal employment sector and the design of social protection.
Despite the long-term effects of the 1949 reforms on the design of the welfare state, the 1950s did not experience a tangible upswing in social spending, and support for people out of work remained feeble, as the government commanded insufficient income streams from taxation (Henderson, 2001, p. 406).In fact, social security in Colombia was "self-financed [receiving] no government transfers" and covering roughly 14% of the labour force in the latter part of the 1950s (Berry & Urrutia, 1976, p. 150).In other words, spending on welfare did not usually extend beyond the amount of contributions made by formal employers and formal employees via payroll taxes as the welfare budget was unlikely to receive cross-subsidies from general tax revenue, so that the "social security system as a whole probably [did] not redistribute income in a significant way" (Berry & Urrutia, 1976, p. 150).
Meanwhile, other Latin American countries had decided to embark on a different policy path of higher public spending (Henderson, 2001, p. 406).Indeed, data from the early-1970s comparing income distribution in six major Latin American economies shows that only Brazil had worse urban inequality than Colombia, whilst Colombia fared worst out of the six countries sampled in terms of rural inequality (Berry & Urrutia, 1976, p. 41).
In any event, Latin American social protection systems were, and in some cases, continue to be non-universal and regressive in distributional terms, rather than progressive as would be required to bring about a more equal distribution of income.The fact that "more generous [benefits have been] subsidized by general taxes" to which the poor, of course, contribute relatively more than the well-off continues to be an exacerbating factor (Huber et al., 2008, p. 423).
According to Itzigsohn, "social citizenship […] was achieved through work in the formal economy", so that eligibility for social insurance depended on an individual's occupational status rather than the individual's status as a citizen (2000, 153-154;Haggard & Kaufman, 2008, p. 4).However, as the formal economy has not expanded sufficiently, for many years coverage of social insurance has been low and of limited quality, exemplified by the state of health care or pensions.Thus, the empowering and protective opportunities of social citizenship have been hampered by the lack of formal employment and the shortcomings of the social protection edifice.
A system that ties access to social insurance to a formal employment relationship, and according to which the economic factor of labour is taxed highly vis-à-vis capital or land, due to relatively elevated payroll taxes, does not provide an incentive for formal employment creation.Despite the welfare state's insufficient coverage during the ISI era, the populist policies of the then prevailing economic orthodoxy managed to maintain the illusion of expanding in unsustainable fashion social protection programmes until the ISI paradigm eventually collapsed under the weight of its own political and economic contradictions.
In fact, the period from the post-World War I era until the 1980s did not facilitate a progressive advance of citizenship, whose participatory and discursive elements could have informed a rethinking of social and labour regulation as well as employment generating policies in an attempt to adapt to changing socio-economic and political winds in the national as well as the international arena.During the post-WWII era it was becoming increasingly apparent to policymakers that the assumptions underlying economic planning were at odds with reality, as the industrial potential of Latin American economies proved unable to absorb a significant share of the economically active population (Bangasser, 2000 2, p. 4).
Although Latin American countries in many respects did not meet the conditions of socioeconomic and political development that proved to be the prerequisite for welfare state creation in Western Europe, the early Latin American industrializers decided to broadly follow the style and function of the European model.One can assert that the payroll tax-financed or occupationallybased social security system originated in Germany in the late 19th century under Bismarck's political leadership.
One of the reasons why the system worked relatively well in Germany and subsequently in other Western European countries was an already active and established labour movement, which became increasingly organized with the advancing process of industrialization."While in the late 1890s collective agreements hardly existed in Germany, by 1913 [they covered about 20 per cent of the total manual work-force]", and organized labour maintained such accords with threats of strikes (Clark, 1983, p. 83-84).Furthermore, most employment tended to be created in the formal sector.
A number of German social analysts and policy-makers initiated regular meetings in the Verein für Socialpolitik (Social Policy Association) in 1873, "which is generally seen as a precursor of modern social policy thinking" (Verschraegen, 2009, p. 148).It is nevertheless revealing that the first steps towards modern social legislation in Germany were only undertaken after a 12-year period of repressive state action against labour movements (Pappi, Koenig, & Knoke, 1995, p. 73), during which the executive endeavoured "to outlaw socialism in 1876" (Henderson, 2001, p. 4).
Social insurance became enshrined in law without the official participation of labour movements, whose desired policy goals were, however, "presumed" by the government and business associations (Pappi et al., 1995, 73, my translation).Around the same time, the working classes across Europe developed a consciousness of the factors contributing to "their exploitation by the moneyed classes" (Henderson, 2001, p. 4).Labour parties formed in Sweden and Britain in 1889, and unions demonstrated their powers of mobilization during "the great London dock strike of 1889" (Henderson, 2001, p. 5).
An additional factor that facilitated the consolidation of the European welfare model was the evolution of the employment contract as a legal institution.Developments in labour law established the "contractual foundation" of the employment relationship regulating the "essential rights and obligations" of employers and workers (Kahn-Freund, 1967, p. 635).Similarly, Deakin and Wilkinson argue that the legal institution of "the contract of employment [joined] the enterprise to the welfare state" (2005, p. 17).
In the case of Colombia, however, the history of the individual employment contract began with Law 10 in 1934 (Guataquí, 2001, p. 174).That law detailed the obligations of both parties to the contract (Article 13); and stipulated fifteen days of paid holidays as well as sick benefits for up to one hundred and twenty days (Article 14).Claims to severance pay whose level depended on the total numbers of years of service (Article 14); and Article 15 prohibited workdays exceeding eight hours.Nevertheless, a further step towards facilitating the actual implementation of Law 10 did not come about until 1944 when a presidential decree established a separate judicial structure for labour cases (Rodríguez, 2001, p. 616).
Moreover, political revolutionaries who had travelled to Europe and journeyed across Latin America, as well as print media and the arrival of waves of European immigrants, inspired the spread of socialist ideas across the region, thereby increasing the pressure on the local ruling elites, who reacted by hesitantly attending to labour's demands by introducing mostly limited social reforms as evidenced by the cases of Chile and Colombia.However, similar to the spread of industrialization, such efforts materialized with a degree of time lag between what could be considered the pace-setting events in some European countries and the coming into being of socio-economic change for some sectors of society in Latin America.
Among a number of reasons for adopting ISI were the meagre experiences several Latin American countries had had since independence as exporters of primary products that over-relied on the "commodity lottery" rather than on competitive exports industries (Bulmer-Thomas, 2003, p. 17).The subsequent demise of the ISI political economy doctrine led to an increase in unemployment and informality that eroded the already strained contributory basis of social security, decreased the number of potential recipients due to a fall in formal employment, and resulted in "a reduction of […] social security expenditures" as well as moves towards private forms of social security, especially in the area of pensions (Huber et al., 2008, 423, 432;Huber & Stephens, 2012, p. 173).During the Latin American debt crisis of the 1980s the dominant policy choice of structural adjustment imposed "economic constraints [on governments' attempts] to expand the general schemes to cover the growing informal sector" (Huber et al., 2008, p. 432).In a nutshell, the Latin American welfare state was truncated, assuring "some form of stratified social citizenship" (Filgueira, 2007, p. 143).
The rapid and disorganized waves of urbanization in Latin America, as well as the resultant and continuing pressures on urban labour markets, only further complicate this picture.During the Cold War era, the process of incorporation was often expressed as the coercion of organized labour "into relatively docile […] organizations [being] fully integrated, under the watchful regulatory eye of the state, into […] their respective societies" (Bergquist, 1986, p. 4).In other words, workers gained more formal de jure rights, but they lost de facto political power along the way.
Thus, political coercion of the labour movement and its political allies "led progressively to the breakdown of open political systems", which, of course, directly restricted the number of political interest groups attempting to influence labour and social regulation (Bergquist, 1986, p. 3).Yet, according to Bergquist's mid-1980s analysis, it is important to consider that "organized labour is central to the continuing post-war crisis in the main countries of the [Latin American] region" (1986, p. 7).
Moreover, in most Latin American societies mainly export-led capitalist economic development underwent a rapid expansion during the late 19th century, so that an organized working class began to form (Boris, 1990, p. 13).However, exceptions existed as not all of the economically important Latin American countries conformed to that pattern of industrialization coupled with formation of a working class.We learn from Henderson, for example, that in Colombia during the late 19th century "[t]here were no factories, no labour unions, no socialist parties, no urbanization" (2001, p. 5).It was only in the mid-1910s that "Colombia's labour movement experienced considerable growth" evidenced by the publication of "Socialist newspapers" and strike action.
According to Bergquist, "[i]t is these workers […] whose struggles most deeply influenced the modern trajectory of the various national labour movements of the region" (1986, p. 8).Labour only became a relevant factor in the political life of Latin American countries during the 1930s and 1940s when economic policy-making began to be increasingly geared towards industrialization and to cater to the domestic market (Boris, 1990, p. 13).However, the labour movements' organization was not reflected in achieving correspondingly sizeable amounts of political clout, which has been explained by state sanctioned violence and the heterogeneous structure of the working classes (Boris, 1990, p. 13-14).
The existence of a considerable informal employment sector was a contributing factor to that state of heterogeneity.Informally produced goods and services tended to undercut the prices of corresponding products originating in the formal economy, and it tended to be more profitable for owners of capital to exploit loopholes, paying lower wages to informal workers than to their formal peers (Boris, 1990, p. 32).The state maintained a strong grip on labour relations and "[i]n most of the countries of the region […] the trajectory of national labour movements established by midcentury has held firm in our own times", affecting politics throughout the region in the post-World War II era (Bergquist, 1986, p. 8).Public sector unions, especially, grew in importance during that time, leading to what Dombois and Pries refer to as "corporatist or clientelistic entanglement" (1999, 34, my translation), or "cozy relationships […] undermining [the state's] coherence" (Malloy, 1993, p. 223).
Historically, Latin American governments "have shown an obsession for the political control of organized labour and its actions" (Ermida-Uriarte, 1993, 31, my translation).The zeal for government intervention has by no means been limited to the treatment of unions, but has also extended to the making of detailed requirements for the design of employment relationships such as in terms of minimum wages, working time, compensation for dismissal and social benefits (Dombois & Pries, 1999, p. 34).The ISI period nation-state and its predecessors placed a hefty burden "on the productive structure of society", withdrawing far more funds from the economy than the state added "as a deployable resource for public policy" (Malloy, 1993, p. 223).In a nutshell, statesociety relations took on a paternalistic and clientelistic character rather than being defined in terms of rights and citizenship as advanced by Marshall (1950) and Bendix (1964).
Rather than being a result of political action from below, the state orchestrated large parts of the labour market as part of an overarching industrial policy, which would over time come to influence sectors of the economy that were not as intensely controlled by consecutive governments (Dombois & Pries, 1999, p. 34).Thus, at least some of the supposed and actual gains by worker organizations can better be characterized as attempts of incorporation originating from the state, or "[binding] urban labour to the state […] via a regime of state-controlled trade unions and welfare enhancement" (Lewis, 2019, p. 14), rather than being a result of years of political strife.
At the same time, one must stress that unions and their members suffered decades of hardship during oppressive military regimes, lasting until the mid-to late-1980s.In more recent events, former union leaders and labour activists were propelled to the presidency with the political support of organized labour, which was the case in Bolivia and most prominently in Brazil with the election of Luiz Inácio Lula da Silva serving two consecutive presidential terms between 2003 and 2010.Lula da Silva, however, had run for president on five occasions before eventually becoming head of state and head of government (Rathbone, 2012).
Finally, the peripatetic progress of incorporating labour and the varying degrees of political mobilization over time demonstrate the stark contrast between the formal and the informal dimensions of the labour market (Bromley & Wilson, 2018), which the intensifying social citizenship discourse in Latin American societies, and the resultant attempts at redefining, constitutionalizing or extending social rights, will set out to reconcile in times to come.
Due to the size of its territory, population and natural resources, Brazil has been among the largest and most influential economies in Latin America and also among the region's early adopters of social security and labour regulation.Other countries in the "pioneer group" were Argentina, Chile, Cuba and Uruguay (Mesa-Lago, 1994, p. 15).Similar to the Argentine case, the depression of the 1930s stimulated the growth of "domestic industry and an urban working class" in Brazil (O'Donnell, 1973, p. 54).
Apart from a social security system for public employees founded in 1926 prior to the Estado Novo, the first administration led by Getúlio Vargas (1882Vargas ( -1954) ) between 1930 and 1945 spurred the Brazilian welfare state with unprecedented progress in terms of depth and coverage (Baer, 1965, p. 96).Thus far, one could presume that at a basic level Vargas' policy agenda appeared to be benign towards those generally situated at the less influential side of the employment relationship.However, the political factions taking power in Brazil with Vargas as their figurehead were "a peculiar [and technocratic] coalition of landholders and others […] representing a crucial step […] toward greater penetration […] of society by the central state" (Evans, 1979, p. 86).
The main thrust of the subsequent brief analysis deals with the Estado Novo period from 1937 until 1945.This era was characterized by relative political and economic stability under an authoritarian political system that sought to bring about economic development.This state of relative calm contrasted with periods of political turmoil both prior to and during the aftermath of the Estado Novo (Liu, 2006).Among the key features of the Vargas administration was "[i]ntervention by means of regulation" and a focus on "state entrepreneurship" as a vehicle of economic development (Evans, 1979, p. 87).In a comprehensive study of citizenship and modernity in Brazil, Holston refers to the Estado Novo as a "corporatist and populist state" bringing to pass a "perpetuated [version of] the nineteenth-century paradigm of differentiated citizenship" in the form of "state-sponsored citizenship that has lasted in part through the contemporary period" (2008, p. 186).Furthermore, Evans adds that "[c]orporatism was […] an ideology of social control [in order to] integrate Brazilian society from the top down " (1979, p. 86).An exemplary case of the lasting effect of the social and labour legislation enacted under the Vargas regime is evidenced by the publication of earnings data of public employees working for the São Paulo municipal assembly revealing that in 2012 "a nurse in charge of a drop-in clinic for municipal workers […] earns […] 12 times the average private-sector wage [and] a third of ministers and nearly 4,000 other federal employees breach the [constitutionally set] pay ceiling" (The Economist, 2012).Thus, an analysis of the social and labour legislation of the Brazilian Estado Novo throws light on the evolution of citizenship with special attention to the labour market that is still relevant for one of Latin America's dominant countries today.
Furthermore, an understanding of the underlying structures of the CLT provides insights into the elite's attempts at incorporating the urban working class into the state's power system via rules of dispute settlement that favoured "impersonal rules [over] personal negotiation" (Erickson, 1972, p. 146).In other words, the CLT enforced a dispute resolution mechanism that intentionally eschewed political struggle in the form of competitive politics under the participation of opposition parties as well as trade unions, and instead institutionalized a politically charged and biased system of labour courts and political patronage.State-orchestrated wage negotiations and labour disputes via the court system made collective bargaining unattainable (Baer, 1965, p. 96).Put differently, rather than creating the necessary policy space for privately conducted negotiations between employers and workers, the state structured disputes on wages and working conditions via "public controls" (Cox, 1997, p. 127).The Vargas regime hence "maintained control over the working classes", which also gained it approval from Brazilian industrialists (Evans, 1979, p. 90).
According to Erickson, the CLT is a "highly structured, minutely regulated labour code" (1972, p. 146) which at the same time appears to be "quite elaborate and advanced for an underdeveloped country" taking its first steps towards industrialization (Baer, 1965, p. 96).Baer's comments on the state of Brazilian development can be interpreted as an academic understatement, for one could refer to 1930s Brazil as "semi feudal" with a largely illiterate working class and very prosperous landowners (The Economist, 2012).In any event, the Vargas administration created a total of 38 individual funds covering workers in different industries providing them access to health care, pensions and vocational training colleges, the latter of which were part funded by workers contributing 1% of their earnings (Baer, 1965, p. 96-97).That degree of complexity within the public social insurance system indicates that certain professional groups received preferential treatment, which further divided society, rather than creating a focal point of citizenship formation.What is more, the system's many layers could easily overburden the administrative capacity of the state.
That activist period of Brazilian welfare state creation saw the inception of strict dismissal laws and minimum wage legislation (Baer, 1965, p. 97-98).Despite having the appearance of a progressive social policy programme, the introduction of the CLT did little to change the fact that "even in 1964 less than 40 per cent of the [urban] labour force [was] covered" and levels of implementation could not be reliably measured (Baer, 1965, p. 98).
Thus, the overview of the CLT's properties supports the conclusion that the system represented a top-down, state-sanctioned policy that differentiated and subsequently fragmented workers whilst failing to equip them with the tools to take an active part in the politics of the labour market and social insurance.In other words, it is certainly questionable whether the Estado Novo approach to solving the social question, industrialization and urbanization resulted in workers receiving relief as a right instead of as an issue of charity.
Regarding the key motivation driving the Vargas administration and the military regime usurping Brazilian politics in 1964, Evans highlights "the necessity of making the revolution from above before someone else succeeded in starting one from below " (1979, p. 93).The Estado Novo, as well as other political regimes in Latin America, did not provide much policy space for non-state actors to participate in policy-making and implementation.Vargas' industrial relations policies that enabled merely truncated forms of labour organization "under the wing of the state" provide us with a succinct example underlining that claim (Evans, 1979, p. 48).
Offering a glimpse into the future of social and labour regulation in Brazil after the Estado Novo, approaches to the study of the CLT from the perspective of the constitutional function of labour law provide an additional insight.The Vargas regime's unwillingness to apply "[l]aw […] to recognise a system of workers' councils and bi-partite industrial councils, and to confirm the continued existence of the trade unions and employers' associations", as was mandated by the 1919 Weimar Constitution in Germany, for example, seems to have set Brazil's labour and social regulation on a perilous institutional path (Dukes, 2011, p. 62).
Put differently, the labour and industrial relations regime codified during the Estado Novo did not set in motion legal developments that would over time limit the role of an activist state, bring about shifts in the relationship between capital and labour, introduce elements of political competition into trade unions and business associations, as well as advance political participation and political bargaining.It can be argued that as a critical juncture (Collier & Collier, 1991) in the social history of Brazil, the social and labour regulation of the CLT decidedly contributed to putting a firm lid on the development of alternative, more immediate and potentially far more transformative paths of institutional change (Touraine, 1989).
The realities rather than the rhetoric of the CLT also point to deep-seated asymmetries in societal power relations, and to the analytical theme of the stickiness of institutions vis-à-vis the capacity for incremental, endogenous institutional evolution.One would expect that the Brazilian economy's transition away from agriculture towards industrialization unleashed forces impinging on the fabric of society and setting in train structural change.In spite of the transformative power of industrialization on overall society, the impact on politics was relatively muted.
Whereas the rural economy had by and large adopted and perpetuated the stark power asymmetries locked-in during the colonial era, urbanization and the accompanying rise of the working class vying for political participation subsequently threatened to shake the societal status quo to the core.It appears that political and economic elites accepted the fruits of the emerging manufacturing sectors' output and the ensuing availability of certain luxury imports, whilst staunchly resisting institutional change of the socio-economic and political make-up of Brazilian society.Yet Lewis reminds us that "the post-1970s industrial expansion of countries like Brazil and Mexico [could likely not] have been possible without the manufacturing platform established during the classic age of [ISI] between c.1945[ISI] between c. and 1972[ISI] between c. /3" (2019, p. 22), p. 22).
Thus, the opportunity to encourage the insertion of wage labour into the labour market coupled with social policies such as social insurance and vocational training programmes was largely missed, leaving many workers uninsured and at risk of labouring in the unprotected, informal sector (Deakin, 2011, p. 165).Even though Vargas undoubtedly pushed through reforms to alter the structure of Brazil's economy, the social and labour legislation enacted under his rule hampered a restructuring of socio-economic relations.
Finally, it can be argued that the Brazilian "mobilising state" (Touraine, 1994, p. 46) undermined the relevance of social conflict with some degree of success.Despite the persistence of the institutional path that the CLT forcefully established, the aforementioned clearly highlights the presence of discursive patterns within societal relations and points to the potential for social conflict that could challenge the structural forces upholding the prevailing institutional settlement.

Conclusions
This paper highlights the centrality of debates on society's responses to the processes making up industrialization in Latin America and attempts to ease the inherent tension between the State and the market.Historically speaking, Latin American social protection regimes and labour market institutions tie in with the region's colonial past, as well as with the strong influence that European ideas and practices have had on the incorporation of labour into society since the onset of industrialization and the rise of the welfare state.The Marshallian analysis of citizenship as featuring the triad of civil, political and social rights indicates that the road towards a European welfare state built on universal principles was rather more error prone than straightforward.Western European attempts were marked by class struggle as well as oppression and exploitation during the early stages of industrialization.The ISI state in Latin America was comparatively strong in the political and socio-economic spheres, which consequently weakened the status of citizenship.
Crucially important for social rights in Latin America is the temporal dimension of the onset of industrialization.The significance of this for the development of social rights in Latin America lies in the fact that political and civil rights as well as the process of industrialization were either at a nascent stage or had not properly taken hold when local elites began the project of welfare state creation.In the context of economic development, the notion of the welfare state evolves from "[reacting] to the effects of industrialisation with measures of social help" towards '[meaning] more than just social help and more than just compensation for injuries' (Luhmann, 1990, p. 21).In fact, Latin American industrialization was not accompanied by the "formation of an industrial working class and labour movement" and hence tended to lack a crucial emancipatory force conducive to citizenship (Huber, 1988, p. 23).A coalition consisting of 'the State, the protected industrial bourgeoisie, and their workers' (Kaztman & Retamoso, 2005, p. 132), was motivated by political rather than capitalist economic motives.