https://cpes.org.uk/om/files/original/ddb09960e3b7c32a1147175cae36e5d8.pdf 95f680b975a9222937a2b563b7899605 Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource International Contributions to Labour Studies 6 Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Text Any textual data included in the document International Contributions to Labour Studies (1996) 6, 29-48 Weakening employment protection in Latin America: incentive to employment creation or to increasing instability?* Adriana Marshallt Introduction At the turn of the 1980s and during the early 1990s, a number of Latin American countries weakened their legal systems that protected labour in the private sector. While several of these reforms specifically addressed the employment protection regime, in some countries the mechanisms of wage determination and collective labour rights were also revised. In Argentina (1991), Colombia (1990), Ecuador (1991) and Peru (1991), for instance, forms of 'flexibilisation' of the employment contract and reforms to the legal protection of unfair and collective dismissal were implemented. In contrast, in Brazil (1988), Chile (1990-1991) and Venezuela (1990), labour legislation was modified so as to improve employment protection and worker rights. Other Latin American countries, such as Mexico, did not join in this process of labour law reform, but introduced some changes to labour protection through collective agreements. In these years also, 'state reforms' were redefining the scope of state intervention and therefore the size, composition and labour market role of public-sector employment. The aim of this paper is to discuss labour market and employment impacts of the employment protection reforms on private sector workers. While the main concern lies in the effects of curtailed protection, countries where worker rights were strengthened are also considered; this parallel analysis helps to grasp the employment consequences of legal reforms. 'Employment protection' (a term used very loosely throughout this paper to refer to the legal institutions that place barriers on the free utilisation of individual and collective dismissals and of less or unprotected employment contracts) has been *This article is based on research supported by an award from the Centre for International Studies of the University of Toronto. tConsejo Nacional de Investigaciones Cientificas y Tecnicas and Instituto de Desarrollo Econ6mico y Social, Buenos Aires. 1052-9187/96/010029+20 $25.00/0 © 1996 Academic Press Limited 30 A. Marshall viewed as both detrimental and favourable to employment performance. 1 Succinctly, it has been argued that protection intensifies unemployment and inhibits employment growth by stimulating labour substitution even in labour surplus situations, by deterring employers from recruiting at times of expansion and by fostering segmentation between the employed and the unemployed, and that it encourages the utilisation of temporary contracts and overtime working. In contrast, it is argued that by placing constraints on dismissals during recessions, it helps to restrain unemployment growth and to stabilise labour demand in the longer term.2 Furthermore, employment protection has been perceived as being both adverse and beneficial to productivity growth,3 often selectively emphasising the impacts on aggregate productivity via individual and collective worker behaviour, or instead through employer practices: adverse, because (from diverse theoretical stands) it is supposed to moderate the fear of job loss, thus relaxing work effort and labour discipline, and to strengthen workers' collective resistance to technological and organisational changes that save labour, require a workforce with different skills and characteristics or affect work arrangements; positive, because protection would encourage labour-saving innovations, help to secure workers' commitment to enterprise success, induce employers to provide and workers to acquire firm-specific training and skills, and favour co-operative relations at the workplace. However, it is also suggested that the outcomes of protection on productivity depend on the conflictive or co-operative nature of industrial relations and on labour market conditions (Buchele and Christiansen, 1992; DeFreitas and Marshall, 1995). After more than a decade of intense dispute, the debate on the impact of employment protection regimes on the labour market has not yet been settled. In relation to the employment effects-the focus of this paper-it was not possible to demonstrate as yet that stronger protection against job loss and limitations to flexible contracts lead to less employment creation in the long run or the opposite. Contradictory findings can be called on to support either position.4 Now we are aware that the problem is much more complex than the usual simplifications posited by both sides, that the effects of employment protection vary according to social and economic circumstances, and that the many institutions that contribute to employment protection, apart from laws, play a role in determining labour market outcomes. Evidence shows, nonetheless, that the legal regime of employment protection (consisting ofa set ofinstitutions that vary across countries) does have an effect on employer and worker behaviour, and that this structures the labour market. In other words, the imposition ofinstitutional restrictions on the managerial prerogative to 'adjust' the workforce to changing business conditions, as well as their 1 I will not refer in what follows to the unquestionably beneficial social effects of protection nor to the several controverted global economic consequences, except for the closely interrelated impacts on employment and productivity. 2 There are a number of variations around these basic contentions. For a detailed account of the diverse employment effects attributed to job security, see, for example, Emerson (1988) and Buechtemann (1993). 3 Inter alia Weisskopf (1987), Rebitzer (1987) and You (1992). Of course, in certain cases there is a trade-off between positive effects on productivity and on employment. For a critical discussion of the contradictory productivity effects attributed to employment protection, see Marshall (1994a). 4 This evidence is reviewed in detail by Buechtemann ( 1993). Employment protection in Latin America 31 degree of stringency, influence the structures of employment and unemployment, although not necessarily employment and unemployment rates and trends. However, while in relation to OECD countries there has been an increasing recognition (at least in academia) of the lack of a straightforward relationship between employment protection regulations and labour market global performance, many authors concerned with Latin American countries still assert, on frail empirical grounds, that only if the 'obstacles' and 'rigidities' that come from employment protection were to be removed, would jobs expand and productivity improve. 1 Nonetheless, a comparative study of a number of Latin American countries with considerable variation in the extent to which the entrepreneurial prerogative is restricted found that the laws regulating contracts and dismissals, by affecting both employer practices and the behaviour of wage earners, structure the labour market but have no discernible effect upon the performance of macroeconomic variables (employment, productivity, exports), determined mainly by other factors (Marshall, 1994a). This paper endeavours to contribute to the wider debate on the labour market effects of employment protection regimes. It presents an empirical analysis of the employment impacts of labour law reforms via changes in employer practices, looking at individual countries over time.2 To this end, I discuss the reforms to the employment protection regime for the private sector implemented c. 1990 in seven South American countries (Argentina, Brazil, Chile, Colombia, Ecuador, Peru and Venezuela) that dealt with employment contracts and individual and collective dismissals, and an attempt is made to identify some of their short- and mediumterm labour market outcomes. I contend that, if it is true that the 'flexibilisation' of contractual-dismissal rules has an influence on employer dismissal and recruitment strategies, this would be reflected by an increased sensitivity of employment to variations in output (i.e. increased employment-output elasticities and turnover rates) and by changes in the structures of employment (e.g. in the proportion of temporary workers and of workers with short-term job tenure) and of unemployment (this latter not discussed in this paper). Furthermore, it is argued that whether these effects, in turn, result or not in higher employment levels or faster employment growth is a strictly empirical question: short-term increases in both dismissals and recruitment may or may not balance out in the long run. Obviously, legal employment protection is but one, and surely quite a secondary, determinant of employment growth and the employment rate. Given output trends determined inter alia by the domestic and external demand and by investment rates, the degree of employment creation depends basically on changes in the economic structure and investment patterns, on technological and organisational innovations, and on the 1 Primo Braga et al. (1995, p. 22), for example, view the growth of unemployment in Argentina following trade liberalisation as the outcome of 'onerous labor market regulations [including] antiemployment accidents legislation, collective bargaining set at the level of the industry ..., and high and distortive taxes on wages', and, in relation to Peru, they assert that '[labor security legislation] reduced drastically the creation of productive employment opportunities in the formal sector'. In a similar vein, Nairn (1995, p. 58) stated that '[Latin American) exporters still face such hurdles as labor codes that stifle worker productivity.' 2 Therefore, in this study, other effects of employment protection, such as those on productivity, are not explored. 32 A. Marshall ensuing growth rate of productivity; the availability and cost of labour and the constraints on its use and deployment have some influence on those processes. The paper is organised as follows. Firstly, I review labour law coverage and some relevant aspects of employment protection regimes in the selected South American countries to assess how these countries stand vis-a-vis the industrialised OECD countries and the significance of the labour force segment protected by legislation. Then, the rationale behind labour law reforms and the specific changes introduced are discussed. Finally, some of their labour market effects are analysed in detail, concluding that the evidence suggests that the weakening of employment protection, while likely to have influenced employer practices and short-term employment changes in the foreseeable direction (expansion of dismissals and of job instability), did not stimulate employment creation in the medium term; thus, the declared purpose of the reforms was not fulfilled. 1. Employment protection compared In the South American countries examined here, the labour code protects, in principle, all wage earners employed in private activity, 1 and, generally, in stateowned enterprises, while civil service employees are covered by special statutes. The size of the labour force segment protected by the labour code thus varies across countries according to the importance of wage in total employment. At the turn of the 1980s, waged workers accounted for over 60% of the total labour force in Argentina, Brazil, Chile and Venezuela, and over 50% of the urban labour force in Colombia, Ecuador and Peru. 2 State employees protected by the special civil service regulations comprised some 15-25% of all wage earners (Marshall, 1990). Many wage earners are employed not or only partly complying with labour laws. These 'informal' waged workers represented about one third of wage employment in important cities of several of the countries analysed in this paper. 3 They receive few or none of the social benefits stipulated legally; obligations concerning social security contributions and lay-off compensation tend to be evaded. Nonetheless, these workers can prove that they are covered by labour legislation. However, wage earners 'disguised' as self-employed workers (one of the most regularly used mechanisms to avoid payment of social benefits being to circumvent the wage employment status) would seldom be able to provide proof of entitlement to the benefits and are thus wholly unprotected. Still, labour law coverage in Latin America, although less extensive than in the advanced OECD countries where the importance of wage in total employment is much more substantial and enforcement stricter, cannot be said to be negligible, so much so that at the turn of the 1980s the elimination of many, allegedly 'excessive', labour regulations became one of the cornerstones of the economic policy packages. 1 With the exception, in some countries, of special groups, such as domestic service, rural workers and homeworkers, often covered by ad hoc statutes. 2 Data are from population censuses and household surveys. 3 Estimates diverge according to whether one considers either social security coverage alone or obtention of all social benefits. In countries such as Brazil, there are wide differences in coverage across regions. Employment protection in Latin America 33 The (restrictive or permissive) nature of employment protection regimes depends on dismissal laws and also on the legal regulation of employment contracts not protected by lay-off legislation, and on the mechanisms of employment protection agreed through collective bargaining. 1 Legal employment protection regulations (contracts and dismissals)2 prevailing in the Latin American region until the end of the 1980s provided for advance notice and/or lay-off compensation and, in certain cases, worker reinstatement; the just causes of lay-offs were stipulated in detail (typically, to leave work and repeated absence without a valid motive, intentional damage, and serious offence) and seldom included the economic needs of the firms; and temporary contracts faced restrictions in several countries. In some countries, the so-called capitalisation funds, according to which employers contribute to individual workers' employment termination funds a proportion of wages (which means that there is a fixed cost for employers, independent of each individual dismissal), co-existed with or replaced employer severance pay due in the event of unfair and collective lay-offs. Regulations are relatively intricate; this complexity was often the product of successive partial labour law modifications. The countries examined had, prior to the 1990s reforms, either employer dismissal compensation (Argentina, Chile and Peru) or a combination of employer compensation and capitalisation fund (Colombia, Ecuador, Venezuela and to some extent Brazil). Not all of them included in their labour codes the requisite of advance notification in the case of unfair dismissal, and divergences across countries in the regulation of collective lay-offs were noticeable (Chile and Venezuela, for example, having no restrictions). The reforms (examined below in detail) led to greater convergence of employment protection regimes because countries with tighter regulations tended to relax them while those with weaker protection enhanced it. Besides, additional countries established capitalisation funds (Peru). All of these countries had regulations on the use of temporary contracts, and, in this area too, greater convergence took place after the reforms of the 1990s.3 Employment protection regimes in Latin America have usually been regarded as being fairly extensive according to international standards, and highly protective of workers' rights against dismissal. The comparison with EEC countries highlights, however, the limitations of South American employment standards legislation. In most of the EEC countries, the law provides for advance notice and compensation in case of unfair dismissal (length and amount varying across countries), and in many of them temporary contracts are regulated (only two of the 12 EEC countries have no restrictions whatsoever, and a further three have few limitations). 4 Collective agreements may include clauses on massive dismissals and temporary 1 Even though non-compliance with labour laws generally is considerable, the differences in the objective possibilities for potential legal suits following unfair practices, deriving from cross-country variations in employment protection regimes, should be reflected in differential degrees of effective labour protection. 2 Regulations reached through collective bargaining will not be discussed. Besides, here, as in the rest of this paper, I refer to regulations that apply to the generality of wage earners and not to the many specific norms for women and minors, apprenticeships, and special occupations, professions and areas, etc. included in most labour codes. The only specific cases considered are temporary contractual modalities. 3 Regulations on dismissals and temporary contracts in the selected South American countries are summarily described in Marshall (1996, Tables 1 and 2). 4 Details are in European Commission (1994). OECD (1993) and Mosley (1994). 34 A. Marshall contracts. Besides, before collective lay-offs (and in a few countries also before unfair dismissals) some intervention of work councils or trade unions is very often mandatory, and sometimes prior government authorisation may also be required.' In fact, European employers ranked advance notice requirements and the legal procedures involved in lay-offs as the most serious interferences, above dismissal costs (Emerson, 1988), with their freedom to separate 'at will' unwanted workers. In contrast, in the Latin American countries examined here, some form of government authorisation is required only in the case of collective dismissals and, even then, only in two countries (Colombia and Peru) out of the seven.2 Still more significant is that, although of course Latin American unions often take actions to prevent, delay or minimise lay-offs, particularly in the event of collective dismissals, legally stipulated forms of trade union intervention in negotiating scope, mechanisms and timing of lay-offs, or compensatory plans, are practically non-existent.3 Even the comparison with the US and Canada, where employment protection is much less stringent than in Europe, reveals some of the weaknesses of employment protection in Latin America. In Canada, for instance, severance pay is strikingly lower (2-day wages per year of service and only after 1 year of continuous employment with the firm according to federal regulations) 4 than in the South American countries, but the notice period is substantially longer in the event of collective lay-offs (and, in some provinces, also in cases of unjustified dismissals), and efforts to minimise massive job losses in consultation with worker representatives are often expected (Labour Canada, 1993-94). Lastly, in the South American countries, although legislation on dismissals stipulates the minimum standard of protection, this tends in practice to operate also as a maximum standard (with a few 1 In the Netherlands, individual dismissal requires prior government authorisation, and in some EEC countries employers must consult with employee represesntatives before unfair dismissals; in France and the UK, consultation is required in relation to redundancies. In most EEC countries, collective lay-offs demand prior consultation with worker representatives and advance notice to the government, and prior government authorisation in the Netherlands, Greece (if numbers to be dismissed exceed limitations) and Spain, and also in Portugal until 1989 (see Emerson, 1988; OECD, 1993, 1994; Mosley, 1994); by 1994 the Netherlands was considering abolishing the requirement for government authorisation (European Commission, 1994). In Germany, for instance, the works council must be consulted before dismissals, although management retains the final say; in the event of massive lay-offs, management should inform, consult and negotiate with the works council a social plan to compensate dismissed workers (see Abraham and Houseman, 1993a). Buechtemann (1993) cites evidence on the weak real impact of either works councils (Germany) or government intervention (earlier in France) in impeding dismissals but also notes the psychological effect of regulations. In any case, protection afforded until the early 1990s in Europe in this respect is clearly superior to that which prevails in South American countries, particularly considering the role given to trade unions in Europe. 2 Bronstein (1990) noted that, in Latin America, protection against collective dismissal is underdeveloped. According to Dombois and Pries (1994, p. 66), in the 1980s '[Colombian] employers sought to circumvent the need for government approval for mass redundancies by concluding apparently voluntary redundancy agreements' to avoid potential rejection or delays; after the 1990 law reform, 'applications for mass redundancies are being made directly to the ministry, where they are being decided rapidly and in the employers' favour' (for example, Sofasa-Renault and Avianca). 3 Exceptions, at least on paper, are Brazil and Peru, where, after the reforms, unions were granted some role in discussing collective dismissals. 4 Most of the employment protection legislation in Canada comes from the provincial governments (Baker et al., 1995); federal and provincial legislation do not supersede each other but apply to different activities. In certain aspects, some provincial laws are somewhat more restrictive than federal legislation, but both anyway place Canada at a relatively low position within the OECD context (data in OECD, 1993). Employment protection in Latin America 35 exceptions, confined to specific economic activities), while in OECD countries, laws tend to establish the floor of rights, to be exceeded by terms more favourable to workers agreed through collective bargaining. The scope of application of labour law is rarely universal, even if we consider only wage earners. In general, by identifying those who are entitled to social benefits, laws simultaneously define those who are to be excluded from access to legal protection. 1 This is achieved by fixing entitlement thresholds, in terms of hours of work, firm size, working days, age, etc. (that originate a legally based form oflabour fragmentation that adds to the well-known social mechanisms that generate segmentation in the labour market). South American labour codes have made extensive use of thresholds, in particular to qualify for the strongest forms of protection (such as the right to reinstatement in unfair dismissals). Already, before the reforms of the 1990s, at least 4 hours a day in Peru and 10 years of service in Colombia were required to attain job security; 3 months of continuous employment with the same firm in Argentina and Peru to be entitled to lay-off compensation in case of unfair dismissal, 1 year in Chile and Brazil, and 8 months in Venezuela; and firm size (defined in terms of number of employees and/or production value) affected entitlement to some benefits in Colombia and Venezuela. Apart from the foregoing, the diverse forms of temporary contracts implied specific exclusions from entitlement, particularly to lay-off compensation (although certain temporary contracts required a reduced compensation at termination, e.g. in Argentina). Exclusions affected a variable labour segment (either marginal or quite substantial); for example, temporary workers comprised a mere 1% of wage earners in Sao Paulo, in the range of 10-15% in Buenos Aires and the main Colombian cities but over 30% in Lima).2 Protective legislation and trade-union negotiated regulations mean additional labour costs (in the present case, mandatory severance pay and advance warning of lay-offs), and, besides, many employers consider that they undermine labour discipline and the work effort; this fosters practices devised to avoid or mask either the constitution of the wage employment status (e.g. contracting out to self-employed workers) or at least the paradigmatic relationship (resorting to situations or contracts characterised by some exclusion from entitlement to benefits entailed by the standard employment relationship). These practices may be built upon exclusions already embedded in the legal system of labour regulation ('atypical' or 'non-standard' contracts) or enter into the spurious or fraudulent use of legal exceptions, including fragmentation of the working day to profit from a less than 4-hour threshold, fictitious divisions offirms to qualify for 'small' firm benefits, misuse of temporary contractual modalities, etc. (Marshall, 1992). The least are the legal 'loopholes' available to firms, or the more restrictive is legislation on their utilisation to evade the standard employment relationship, the most extensive will be the use of downright illegal practices including totally clandestine employment forms. Weak law enforcement makes it possible to extend in practice the exclusions where dismissal protection does not apply. 1 This is the 'selective function' of the standard employment contract, according to Muckenberger (1989). 2 Data are for the late 1980s. See Marshall (1994a) on the role of labour laws in explaining these disparities. 36 A. Marshall 2. Labour law reforms The reforms to labour laws undertaken by 1990 in the South American region followed contrasting paths in on the one hand Argentina and Peru, where they were oriented to curtail protection, and on the other Brazil, Chile and Venezuela, where labour rights were improved vis-a-vis what in Brazil and Chile had earlier been clearly less protection than average for the region. Reforms in Colombia and Ecuador, while also globally directed at relaxing worker protection, were somewhat hybrid, in the sense that in a few areas protection was strengthened. The reforms were either introduced once and for all or resulted from piecemeal cumulative changes that, in Peru for instance, implied an extensive body of new regulations enacted through numerous Executive decrees. 2.1. Rationale Labour law is one of the most crucial of state instruments to regulate the supply, price and conditions of use oflabour as well as social conflict (Cortes and Marshall, 1993), and, as such, its contents tend to be reformed to adapt labour market operation to changes in the prevailing economic model. Labour legislation constitutes one of the clearest expressions of the dominant ideology and state's labour policy, but also expresses the outcome of the political struggle inherent to the legislative process. The legal regime tends to be more stable than other labour policies. This notwithstanding, at certain times of abrupt economic transformation and subjected to strong pressure, legislation may be rapidly adjusted to the new economic and social environment, often giving legal status to existing de facto practices. This seems to have been the case with certain South American labour law reforms at the turn of the 1980s, although the many hesitations and numerous contradictory regulations might be one manifestation of the social and political resistance to revise the legal systems. The governments of Argentina, Colombia, Ecuador, Peru and Venezuela, at slightly different times, were actively devising and applying economic programmes that followed the basic tenet of what was to be known as the 'Washington Consensus': redefinition of state activity (deregulation, privatisation), and trade, financial and foreign investment liberalisation. By the early 1990s the switch of Latin America's economic strategies to neoliberal prescriptions was sweeping. In Chile, those objectives had been accomplished much earlier. 1 With the Plan de Convertibilidad of 1991, President Menem's government liberalised trade in Argentina, deregulated the financial system and the inflow of foreign capital, established the free convertibility of domestic currency, while in fact fixing a low exchange rate to be backed by international reserves, partially privatised the social security system, and sold offmost state enterprises (Marshall, 1994b). In Columbia, starting in 1990 during President Gaviria's term, drastic reforms were introduced, with a view to opening up the economy (elimination of the system of import licences and reduction of protectionism in 1990-1991), eliminating foreign exchange controls (1990), restructuring the public sector and changing the social security 1 See, for example, Paus (1994). Economic restrucruring in Chile started about 1974 under the military government, with price, financial and import liberalisation; social security reform followed, while privatisation was to be a more gradual process. Employment protection in Latin America 37 system (1993).1 An earlier (1984-1988) neoliberal stabilisation plan in Ecuador was followed by the more interventionist stance of Borja's administration, during whose term (1988-1992), however, labour laws were reformed.2 More resolute marketoriented policies were reintroduced as from 1992, with President Duran's new stabilisation plan. Labour law reform in Ecuador thus predates the application of the liberalisation programme. However, at the time of the labour law reform, the government had been keen to attract foreign investment, and in 1991 it applied some measures to this end (Hofman and Buitelaar, 1994); the creation of maquila industries (with fewer restrictions on profit remittances and waiving of trade licenses; de Janvry et al., 1994), where the labour code does not apply, as well as the changes in labour laws of general application, should be viewed in the context of this more liberal attitude toward foreign investment. President Fujimori in Peru adopted quite similar policies from 1990 (Seminario and Galarza, 1993), and in Venezuela it was Perez who, from 1989, attempted to put into practice the same ideas; he, however, faced several obstacles that hindered full implementation even before the political collapse of 1992-1993 (Nairn, 1993). In Brazil, however, the steps towards conversion to a more market-oriented economic programme were more hesitant; the process seems to have been gaining momentum only as late as in 1995. In the countries that, at the end of the 1980s, adopted 'flexibilisation' oflabour laws in the frame of neoliberal economic policies and 'state reforms', the passing of new legislation was preceded and accompanied by debates, often heated, that reproduced the 'flexibility' controversies in Western Europe during the 1980s. In Argentina, for instance, the 'flexibilisation' of employment standards started to be discussed in the late 1980s, and by 1995 the debate had not yet ended. The Executive promoted successive projects, each one in turn considered to have failed and to be insufficient to achieve the sought-after labour 'flexibility'; trade union leaders accepted each successive reform in exchange for such concessions as preservation of the traditional trade union control of the funds of the employer-and-worker-financed health-care scheme. In this group of countries, governments and business claimed that the loosening of labour protection was indispensable to foster competitiveness in an increasingly internationally integrated economy. The allegedly high labour protection was considered to have been 'affordable' only in economies sheltered from international competition and no longer appropriate to the new orientation of economic strategies, which had switched, at least in principle, from import-substitution industrialisation for the domestic market toward export-led growth. It was argued that certain labour rights slackened the work effort, discouraged labour productivity growth, fostered illegal employment practices and deterred business from creating employment. The reduction of labour costs and greater freedom to manage the workforce that would result from curtailing employment protection were expected to promote employment generation-at least this was the declared aim and justification of several legal pieces introducing or facilitating flexible contracts, and in Peru the bill indeed was named the 'employment promotion act'. 1 More details can be found in Reyes (1994). 2 See de Janvry et al. (1994) on economic policies in Ecuador and, in particular, the somewhat contradictory steps taken in 1988-1992. 38 A. Marshall Why did labour legislation reform in Brazil (1988), Chile (1990) and Venezuela (1990) go in the opposite direction? Before the reforms Chile, followed by Brazil, had much less restrictive labour protection regimes than did other South American countries, and the change towards improving protection was associated with these countries' democratisation process. The former erosion of labour rights had been one central component not only of the military's repressive labour policies but also of their economic strategies, and in the case of Chile it had implied the total transformation of individual and collective labour legislation (Paus, 1994). From 1990 the civilian government in Chile undertook modest legislative reforms in the context of decreasing unemployment levels and improving economic performance. In Brazil, too, collective labour regulation had experienced major setbacks, and constraints on dismissals had been relaxed in 1966; the 1988 Constitutional reform was intended to reverse, at least partially, that situation. In Venezuela, the new labour law project had been drafted earlier (1985) and was to be approved by Congress after prolonged discussion, 5 years later, during Perez' mandate; Perez opposed a project that was considered to be inconsistent with the ongoing economic liberalisation process and tried to get Congress to postpone its treatment, but he did not exercise his right to veto a law that was supported by most political parties and the unions (Ellner, 1993). 2. 2 Refonns introduced1 Some labour legislation reforms were intended to expand the 'loopholes' legally exempted from employment protection, such as lengthening the 'trial period', during which workers have no rights to protection against dismissal and sometimes neither to other benefits (not implemented in the reforms discussed here; instead, in Venezuela, the minimum period for entitlement to dismissal compensation was shortened); creating new modalities of temporary, casual and other special nonprotected contracts, or eliminating or reducing restrictions on the utilisation of pre-existing ones; and establishing 'free zones' and 'maquilas', not covered by labour codes, to be governed by distinct and laxer rules. The Argentine Employment Law of 1991 created a number of 'promoted' temporary contracts, and the later law (1995) exempted small firms from certain of the restrictions on the use of temporary contracts formulated in 1991.2 A variety of new temporary contracts was created in Peru from 1991, and in Colombia, loosening of limitations on temporary contacts-that were already extremely easy-was enacted in 1990 (elimination of the 1-year minimum).3 'Maquilas' and free zones were created in Ecuador and 1 Here I describe only major reforms affecting the generality of wage earners, i.e. not those applying to specific groups, cases and circumstances. The examples oflegislative reforms mentioned below are not exhaustive. For a synthetic presentation of pre- and post-reform features in each country, see Marshall (1996, Tables 1 and 2). 2 I will not discuss the 1995 reforms in Argentina as it is too early to examine their labour market impacts. 3 In Ecuador, the orientation of the changes in the regulation of temporary work seems to have been less clear: while the 1-year minimum employment requirement was somewhat relaxed, at the same time a maximum duration was fixed and contracts were declared non-renewable, the nature of temporary contracts was defined more precisely and the requisite of 'objective causes' for casual/occasional contracts was introduced. Pita (1993) argues, however, that with the Labour Code reform, a broader range of contracts became available to employers. Employment protection in Latin America 39 Peru,1and in Ecuador part-time contracts for weekend work with ordinary instead of overtime pay were allowed. All of these were intended to facilitate more flexible employment contracts with more limited entitlement to employment protection and social benefits.2 However, in the same years, other countries tightened the constraints on temporary contracts, consistent with their global move toward strengthening labour protection. In Chile, the maximum period for fixed-term contracts was reduced from 2 years to 1, and in Venezuela from 5 to 3 years in the case of white-collar employees (although skilled manual workers became worse off with the new rules); furthermore, in Venezuela the type of task for which these contracts were to be allowed was defined more restrictively. Apart from establishing exclusions to the general protection against dismissal, lay-off regulations themselves were modified directly by addressing, amongst other things, the causes for fair dismissal, the very existence, amount of and entitlement requirements to lay-off compensation, and the rules on job security in the strict sense (right to reinstatement). In Peru, the causes for fair dismissal were extended to include 'absence of punctuality' and deficient worker performance under certain conditions, and the right to reinstatement was weakened. A reduced employer compensation was to be supplemented by the newly created capitalisation fund. In Argentina, lay-off compensation had been increased in 1989 when the threeminimum-wages-per-year limitation to length of service compensation was eliminated; the 1991 Employment Law reintroduced a ceiling, although higher than the earlier one, as it now was three total collectively agreed usual earnings exclusive of seniority benefits.3 The change of 1989 preceded the economic reforms and was against the 'spirit' of the other labour law modifications. In Ecuador the 'stability clause'-a minimum of 1 year of employment before 'eviction' was to be allowedwas relaxed and advance notice eliminated, but compensation for unfair dismissal was made more favourable to workers, and employees who lost their jobs due to collective lay-offs became entitled to compensation.4 In Colombia, employment stability, until 1990 available to workers with 10 or more years of continuous employment with the same firm, was eliminated (and compensation for these workers slightly raised), the global cost of dismissal was cut back,5 and 'firm modernisation' was included explicitly among the accepted causes for massive lay-offs; however, collective dismissal (now more precisely defined) was made more 1 The labour regime to be applied in free zones and maquilas established that all employment contracts (which are renewable indefinitely) are temporary, remunerations and working conditions freely agreed (with some limitations in Ecuador) and there is no job security. 2 At the same time, the activity of temporary labour agencies and intermediaries in some countries was further regulated, generally to prevent abuses and fraud. In Colombia, for instance, the 1990 law defined the rights of workers employed through labour agencies and stipulated the conditions in which such agencies were to be permitted to provide labour. Also, in Argentina, their activities became more extensively regulated as from 1991, worker rights were made explicit and the joint responsibility of the user firm in relation to all social benefits (eliminated in 1976) was reinstated. 3 From 1995 small firms may alter dismissal regulations (upwards or downwards) through collective agreements, and, furthermore, they are favoured by a shorter advance notice period. 4 According to Pita (1993), the reforms tended to undermine employment security. 5 This was due to the reform of the capitalisation fund (to which workers are entitled irrespective of the cause ofemployment mtermination) that eliminated the clause adjusting the accumulated fund each time wages were raised and changed the rules that governed partial withdrawals from the fund. According to L6pez (1993), the pre-reform regulations had stimulated short-term employment. 40 A. Marshall onerous and the previous right of small firms to pay a reduced compensation was eliminated in the case of unfair (but not of collective) lay-offs. In Venezuela, compensation out of the capitalisation fund, the benefits of which are granted to all workers whatever the cause of employment termination, substituted for employer severance pay. However, Venezuela is best characterised by its reinforcement of worker protection against unjustified dismissal, as 'absolute' job security was introduced in 1990 (right to reinstatement)/ although reinstatement may be replaced by the double normal lay-off compensation, to be paid by employers; lay-offs due to economic and technological reasons were given the same status as unfair dismissals; and an effort was made also at characterising more precisely the fair causes of dismissal. The cost of lay-off was increased in Chile, but this had an impact only for workers who had been employed for over 5 years, as the change approximately doubled the ceiling to compensation (1 month wage per year of service) from 5 months' to 330 days' wages; the earlier clauses that had been intended to prevent labour's collective action were deleted from the fair causes of dismissal. On the other hand, in the event of dismissal due to 'modernisation and economic needs of the firm' workers are now entitled to compensation as in unfair lay-offs. In Brazil, the lay-off 'penalty' on employers was increased in 1988 from the earlier mere 10%, to 40% of the accumulated individual worker employment termination fund; the penalty for collective dismissals was also raised. 3. Labour market outcomes Increased protection against unfair and collective dismissal may be expected to be followed by the decline of employment sensitivity to short-run changes in the level of output (i.e. by the fall of the employment-output elasticity) and by less turnover than before the labour law reforms,2 much more so if at the same time regulations on temporary contracts became more restrictive. Lower turnover rates should also be reflected by changes in the structure of job tenure. In contrast, in the countries where dismissal had been made easier and temporary employment specially encouraged, this may be expected to show in higher employment-output elasticities and turnover rates than had been usual before the labour law reforms. In what follows, I discuss whether the evidence supports these hypotheses by focusing on manufacturing.3 We examine first the main trends in the three countries that tightened employment protection. In Brazil, if the two major pre- and post-reform recessions 1 This right to reinstatement is not applicable in firms with fewer than 10 workers. 2 Ideally, the impact of changes in dismissal rules should be examined looking at lay-off rates, but data are not available. Alien and Labadie (1994) analysed trends in the labour force precentage unemployed because of job loss in Chile, finding that a greater freedom to dismiss resulted in higher firing rates. 3 The relationship between employment protection and employment-output elasticities is examined by, for example, Abraham and Houseman (1993a,b) who differentiate between short (monthly) and longer-term adjustment and consider time lags. Given the limitations of available information, these techniques cannot be used in the analysis of the South American countries; therefore, I compare average employment-output elasticities (i.e. the average of annual employment-output elasticities) before and after the legislative reforms, separating the recession and expansion periods. Employment and output data are available only for manufacturing. In relation to Latin America, Marshall (1994a) found that manufacturing employment-output elasticities tended to reflect intercountry differences in employment protection regimes, while some studies over time (on Chile) came up with mixed evidence (see Alien and Labadie, 1994, and the references therein). Employment protection in Latin America Table 1. Employment-output elasticities, * pre- and post-labour law reforms, selected South American countries Recessions Expansions 41 Argentina Pre-reform Post-reform Brazil (metro) Pre-reform Post-reform Chile Pre-reform Post-reform Colombia Pre-reform Post-reform Ecuador Pre-reform Post-reform Peru Pre-reform Post-reform Venezuela Pre-reform Post-reform 1·13 (1982) 0·35 (1985) 1·01 (1987-1990) 10·60 (1982-1983) 0·47 (1988) 6·18 (1990-1992) 1-31 (1982) 7·44 (1982-1983) - 0·04 (1989)" 1·33 (1993) 0·85 (1982-1983) 0·45 (1988-1990) 1·49 (1992) 0·26 (1983) 0·22 (1989) 1·62 (1993-1994)b 0·52 (1983-1984) - 0·32 (1986) - 0·85 (1992-1994) 0·82 (1984-1987) 0·72 (1989) - 0·18 (1993-1994) 2·40 (1983-1990) 0·38 (1991-1994) 0·02 (1984-1990) 0·33 (1991-1994) 0·40 (1990-1991) 0·13 (1992) - 0·28 (1984-1987) - 0·96 (1991) - 0·57 (1993-1994) 0·23 (1982) 1·65 (1984-1988) - 1·44 (1990) 1·53 (1991-1992)b *Elasticities are calculated as averages, for each economic phase, of annual elasticities. "Fall of output and employment growth. bFormal manufacturing only: 2·19 and 0·87 respectively. Source: Own estimates on the basis of annual data in CEPAL (several years) and INEC, Ecuador. (1982-1983 and 1990-1992) are compared, the employment-output elasticity- still rather high-clearly fell (Table 1). 1 This might have been the consequence of a lower firing rate in response to the increased cost of dismissal. Comparing the expansion phases, employment-output elasticities diminished after the reform (actually becoming negative from 1993). In addition, turnover rates decreased gradually after the 1988 reform (with a slight recovery in 1993).2 Turnover has been customarily high in Brazil, owing to both unconstrained dismissals and proclivity of 1 Note that this is not true if the latter is compared with the elasticity in 1988, the year of the new Cosntitution that reformed labour laws. 2 Average turnover per month was 3·80 in 1988 and 2·73 in 1993 (Amadeo and Gonzaga, 1994). The decline in turnover is reflected by changes in the structure of job tenure, as the proportion of workers with up to 3 months employment fell from 14% in 1988 to 10% in 1992, while that of workers with over 10 years seniority rose from 12% to 15%; data are for Sao Paulo's manufacturing (SEADE-Diesse, 1988-1992). However, the most substantial change in the disrribution of job tenure took place by 1992 rather than immediately following the reforms, suggesting that the 1990-1992 recession and subsequent absence of new hiring also played a role. 42 A. Marshall workers to quit their jobs voluntarily in order to have access to the accumulated income in their capitalisation funds. 1 The fall in turnover rates (while the situation with respect to entitlement to the fund did not change) may indicate that lay-offs became less common after the penalty imposed on employers in the event of dismissals was increased, and that voluntary quitting diminished with the growth of unemployment that accompanied the deep recession and productive reorganisation of 1990-1992.2 On the other hand, employment reductions fell exclusively on workers with 'cartera assinada' (registered workers with access to all the legal benefits), - 10·6% in 1989-1992, while employment of non-registered waged workers, with no benefits, and of subcontracted workers increased (2·6% and 10·8% respectively; Andraus, 1993). This employment restructuring to the detriment of regular and permanent employees, towards workers with a more tenuous employment relationship, might have been one result of the tightening of job protection. No special temporary contracts have been made available in Brazil-as they have in Argentina or Peru-and this might have encouraged the expansion of illegal and allegedly non-wage employment. In brief, the strengthening of employment protection in Brazil seems to have had some influence on business practices. At first glance, the evidence for Chile is also consistent with what might be expected from improvements in employment protection. Employment-output elasticities in manufacturing fell after the legislative reform (Table 1), during a period (19911994) of rapid economic expansion. It is, of course, not, however, possible to attribute such a fall unambiguously to a slowing down in recruitment associated with the rise in the costs of potential lay-offs-as the neoliberal argument runs-and to the somewhat less permissive regulations on temporary contracts.3 The acceleration ofthe rate ofgrowth ofproductivity, determined by other factors, could have played a role too (in 1990-1994 manufacturing productivity increased at an annual rate of 2·8%, above the 1980s' annual average of 0·8%); however, Romaguera et al. (1995) conclude that no intensive process of massive labour substitution was ongoing in those years.4 Finally, in Venezuela's post-reform expansion, the employment-output elasticity was lower than in (only some of) the pre-reform growth phases, but in the recession after the labour reform, and contrary to what could be expected in view of the strengthened protection against dismissal, it increased in comparison to pre-reform downturns (Table 1). Turning now to the analysis of short-term employment changes in the countries that 'fiexibilised' dismissal and contracts, what behavioural patterns are apparent? The growth of manufacturing output after the reforms in Argentina was accompanied by declining employment (Table 1). The use of'promoted' temporary 1 The longstanding role of the Brazilian capitalisation fund in stimulating voluntary quitting was noted by several writers (see references cited in Marshall, 1994a). 2 In 1992 unemployment (5·8%) was almost double the rate for 1989 (data in CEPAL, 1994; six metropolitan areas). 3 According to a study of management reactions to changes in legislation in a sample of 21 firms, legislative reforms seem not to have had substantial effects on business practices; in particular, dismissal decisions were not affected by the somewhat higher lay-off costs (Romaguera et al., 1995). 4 Agacino et al. (1995) indicate that, as from 1989, with the increase of investtnent in manufacturing, a gradual and still uncertain 'technological adjustment' process started, led by the most export intensive industries. Employment protection in Latin America 43 employment contracts was modest;1 it was estimated, for example, to have represented only some 1·5% of all new recruitment in the first half of 1994 (Feldman, 1995). Some 40% of these contracts took place in manufacturing, and, as manufacturing employment was falling, it would seem that the expansion of temporary work was at the expense of open-ended contracts.2 If we now compare pre- and post-labour law reform periods of growing manufacturing output in Colombia, we see that the value of the employment-output elasticity increased (Table 1), accompanying the loosening of protection. Besides, following the liberalisation of fixed-term contracts, the share of temporary employment augmented (1991-92), to drop once again in 1993 to the earlier levels.3 The increase is somewhat surprising, as legislation had already been extremely permissive in this respect, and the change in the regulations on contracts was relatively minor. The proportion ofwaged workers in temporary employment increased more in the small than in the larger firms (L6pez, 1993); this may be explained by the fact that, although on the whole lay-off regulations became less restrictive and dismissal less expensive, small firms ceased to be favoured by a reduced compensation in cases of unfair individual dismissal. The parallel growth of temporary employment and decrease in the proportion of workers with less than 1 year tenure4 suggest that, despite the cuts in dismissal costs, temporary contracts were increasingly utilised for longer-term positions as they would continue to entail advantages over permanent contracts. On the other hand, and consistent with the elimination of the reinstatement clause that had applied after 10 years of employment in the firm and had intensified dismissals to avoid reaching this threshold, the proportion of workers with more than 10 years employment seems to have risen slightly.5 In Ecuador, although the employment output-elasticity was higher in the recessive post-reform period than in the preceding recession (when, atypically, output declined but employment increased), if we compare periods of expansion, it is the post-reform employment-output elasticity that is lower. 6 Besides, since the law on maquilas was passed in 1990 and up to early 1993, 37 maquila industries were established, creating close to 3000 jobs, concentrated in two economic activities: textiles and fishing; most maquilas originated in already existing enterprises that were attracted by tax and tariff exemptions and lower labour protection (Pita, 1993). Lastly in Peru, employment-output elasticities increased in the post-reform slowdown vis-avis the pre-reform recessions (Table 1);7 dismissal regulations were weakened and, moreover, special and temporary contracts actively promoted. During the period of economic expansion, employment continued to fall. The jump in temporary employment, from the already outstandingly high level of about 40% of private 1 In fact, employers considered the temporary modalities introduced by 1991 Employment Law to be still too constrained, as trade unions had been granted some power of approval. 2 No data on turnover rates in Argentina exist after 1988, and the structure of job tenure does not appear to have changed significantly (on the basis of the household survey, Buenos Aires, INDEC). 3 Data are for the three main cities; see Berry and Tenjo (1995). They include the self-employed. 4 Data in L6pez (1993), but note that only 2 years (1988 and 1992) are compared. 5 According to data for 1988 and 1992 in L6pez (1993). 6 Employment and output data for Ecuador cover a shorter period (1988-1993) than in the rest of the countries and do not originate in CEPAL (as the others do) but in national sources (INEC). 7 Employment and output data are not strictly comparable. The former refer to firms with 100 employees and over in Lima, the latter to output at the national level. 44 A. Marshall sector wage earners in Lima to some 50% after the 1991 reforms, 1 points to the fact that most new recruitment was made under special flexible contracts and that temporary workers replaced permanent ones. Putting all the evidence together, some tentative conclusions may be drawn. Firstly, in those countries where a wide range of lower-cost, flexible employment contracts were made available to business, or where the use of such contracts became less restricted (temporary contracts in Argentina, Colombia and Peru; maquilas in Ecuador), these new opportunities were utilised, but much more so where, as in Peru, dismissal legislation, despite the reform, continued to be particularly constricting. Secondly, changes in legislation seem to have had some influence on business firing policies. This is apparently the case if we compare employment-output elasticities in pre- and post-reform recessions (they fell in Brazil and increased in Peru and Ecuador); besides, in Colombia, the elimination of the reinstatement clause after 10 years of employment was followed by a change in dismissal policies that affected the structure of job tenure. Finally, recruitment policies in phases of expansion might have also been altered in some countries. This-with the reservations mentioned above-could have been the case in Chile and Brazil (where employment-output elasticities declined) as well as Colombia (where they became higher). In Brazil, the growth of illegal employment and subcontracting after the reform also points to a change in recruitment strategies. However, in no one of the other countries that flexibilised regulations (Argentina, Ecuador and Peru) did recruitment increase. Irrespective of how well employment adjusted to short-term changes in output, was the overall manufacturing employment performance affected by labour law reforms? To answer this question, we will look once again at the comparative employment and output behaviour before (1981-1990) and after (1990-1994) the labour law reforms, but this time at their average trends (Table 2).2 Given output growth, employment in Colombia fared better after protection was relaxed, but so did it in Venezuela where, in contrast, protection was tightened.3 Except for Chile, employment in the remaining countries fell after the reforms, generally intensifying an earlier tendency, and in Argentina and Peru, and to a lesser degree in Ecuador, this occurred in spite of the growth of manufacturing output. The path followed by economic transformation, and its impact on investment and productivity growth, were, of course, more powerful determinants than were changes in labour market regulation. In Argentina and Peru, for instance, the negative employment performance was primarily associated with the effects of economic reforms (trade liberalisation combined with domestic currency appreciation) and the subsequent restructuring (Marshall, 1994b; Verdera, 1995); in Argentina, certain manufacturing activities improved their productivity performance, while other industries, unable to 1 Data from the household survey, Metropolitan Lima (in Verdera, 1995). Figures on temporary workers include workers in the trial period. Furthermore, between 1991 and 1994 the share of temporary employment steadily rose. 2 In other words, while earlier to assess the short-term adjustment of employment we examined each economic phase's average of annual employment-output elasticities, we now look at the elasticity of each (pre- and post-reform) period's average annual rate of change in output and employment. 3 This might no longer be true if only formal-sector manufacturing employment is considered, but no such distinction was made in CEPAL's data for the pre-reform period. Employment protection in Latin America Table 2. Employment performance before and after the labour law reforms: medium-term employment output elasticities * and employment trendst, selected South American countries 1982-1990 (Pre-reform) EOE Employment % change 1991-1994 (post-reform) EOE Employment % change 45 Argentina Brazil Chile Colombia Ecuador Peru Venezuela 3·14 - 0·11" 0·76 -0·29 0·37c 1·47 0·74 - 2·2 -0·3" 2·6 - 1·1 4·5c -2·2 1·6 - 0·66b - 4·31" 0·55 0·10 - 0·34b - 1·07b 1·76d - 3·5b - 2·8" 3·4 0·3 - 1·0b - 6·6b 3·3d *Employment output elasticity (EOE) of annual average percentage changes in each period. tAnnual average rate of change in each period. ·1982-1988 and 1989-1993 respectively. b1992-1994 except for Ecuador, where is for 1992-1993. c1989-1991. dFormal manufacturing only: - 0·53 and - 1·0 respectively. Source: Own estimates on the basis of CEPAL (several years) and INEC, Ecuador. compete and reconvert, simply closed down. In Colombia, manufacturing employment increased after the reform (although more slowly than output), to fall only in 1994. After 1990 employment in non-traditional export manufactures climbed, compensating for job losses elsewhere (L6pez, 1993). In Chile, on the other hand, while employment progressed faster than in the 1980s, the employment-output elasticity declined after the reform, although this was exclusively due to the employment fall of 1994. In brief, there is no systematic relationship between the orientation of reforms and trends in employment creation. Employment trends in manufacturing say little about unemployment behaviour (Table 3). The unemployment rate is influenced, obviously, by changes in employment in all economic activities and in the supply of labour, which have not been analysed in this paper. In Colombia, for instance, unemployment declined with the contribution of manufacturing and construction (L6pez, 1993), while in Argentina it rose sharply as a result of falling labour demand in manufacturing, construction and public utilities, among other activities, at a time when the labour force was expanding (Marshall, 1994b). A perfunctory inspection of unemployment trends in any case suggests that the flexibilisation of employment protection was not followed by reduced joblessness, nor did enhanced protection induce the growth of unemployment. Conclusions After 1990 employment protection regimes within South America tended towards greater convergence, as the most stringent protective regulations were relaxed (as in 46 A. Marshall Table 3. Unemployment trends, selected South American countries 1986 % Urban unemployment 1990 1992 1994" Argentina Brazil Chile Colombia Ecuador Peru Venezuela 5·6 7·5 7·0 11·2 3·6 4·3 5·8 5·5 13·1 6·5 4·9 6·2 13-5 10·5 10·2 9·3 10·7 6·1 8·9 8·1 5·4 8·3 9·4 9·5 12·1 11·0 8·0 8·9 "Preliminary figures. Source: CEPAL, Balance Preliminar de la Economia de America Latina y el Caribe, 1994, no. 556/557, December 1994 (see table A.4 for details on areas covered by the unemployment surveys in each country and other caveats). the case of Peru), and the weakest were strengthened (Brazil and Chile). Labour law reforms were generally consistent with the broader change in economic strategies that swept throughout the region. The new opportunities provided by the 'ftexibilisation' of employment contracts and the introduction of maquila industries seem to have been used by business to replace permanent personnel, but not to the same extent in all the countries analysed. Lay-off strategies seem also to have responded to the slackening of dismissal rules. Similarly, there is some indication that improvements in protection against dismissal affected firing and recruitment policies, and a hint at an inverse relationship between the availability of legal 'loopholes', exempted from wage employment protection, and the use of subcontracting and illegal employment practices. Finally, as it could be expected, there was no discernible impact of the labour law reforms on the average performance of employment, given output trends. Were the reforms to employment protection legislation that relaxed the constraints on dismissal and facilitated flexible contracts effective instruments for the regulation of the labour market? If, on the basis of the evidence cited, we focus on their immediate impact, the answer would seem to be positive: dismissals intensified and temporary work supplanted open-ended contracts. However, the answer is clearly negative if we examine the reform effects on employment creation-the proclaimed purpose of 'ftexibilisation': (with the exception of Colombia, a case of a somewhat more hybrid reform) the curtailment of employment protection was not followed by a better employment performance and, if anything, the ensuing expansion of dismissals and temporary jobs worked exactly in the opposite direction. However, the discussion of the labour market effects of reforms to labour legislation was confined in this paper to the manufacturing sector. Insufficient information on employment trends for activities other than manufacturing makes it difficult if not impossible to analyse the impact of labour law reforms on business Employment protection in Latin America 47 practices beyond this sector. This, as well as the study of workers' behaviour in response to institutional changes-such as the spread of capitalisation funds, creation of unemployment insurance schemes and new rules governing dismissal and contracts-and of how this moulds the supply of labour, over a longer time period, remains open to in-depth research. Bibliography Abraham, K. G. and Houseman, S. N. 1993a. Job Security in America. Lessons from Germany, The Brookings Institution Abraham, K. G. and Houseman, S. N. 1993b. Job security and work force adjustment: how different are U.S. and Japanese practices? in Buechtemann, C. F. (ed.), Employment Security and Labor Market Behavior, lthaca, NY , ILR Press Agacino, R., Rivas, G. and Wormald, G. 1995. La industria chilena: ajuste, evoluci6n, innovaciones y perspectivas, OIT/ACDI, no. 13, Santiago Alien, S. G. and Labadie, G. J. 1994. Labor market flexibility and economic performance in Uruguay and Chile, Final Report to the Tinker Foundation, June (mimeo) Amadeo, E. and Gonzaga, G. 1994. Inflation and economic policy reform: social impli- cations in Brazil, Paris, OECD International Forum on Latin American Perspectives (draft mimeo) Andraus, A. T. 1993. Precariza~ao do mercado de trabalho na Grande Sao Paulo: 1989-92, IILS-PREALC Seminar, Santiago (mimeo) Baker, M., Gunderson, M. and Horton, S. 1995. Labor flexibility and productivity in Canada: markets, institutions and skills, Toronto (mimeo) Berry, A. and Tenjo, J. 1995. Guessing the income distribution effects of trade liberalization and labour reform in Colombia, FOCAUCIS Discussion Papers, FC 1995-5, University of Toronto, Canada Bronstein, A. 1990. Protection against unjustified dismissal in Latin America, International Labour Review, vol. 109, no. 2 Buchele, R. and Christiansen, J. 1992. Industrial relations and productivity growth: a comparative perspective, International Contributions to Labour Studies, Fall Buechtemann, C. F. 1993. Introduction: Employment security and labor markets, in Buechtemann C. F. (ed.), Employment Security and Labor Market Behavior, Ithaca, NY, ILR Press CEPAL (Economic Commission for Latin America and the Caribbean), Estudio Economico de America Latina y el Caribe, several years CEPAL (Economic Commission for Latin America and the Caribbean). 1994. Balance Preliminar de la Economia de America Latina y el Caribe, nos. 556/557, December Cortes, R. and Marshall, A. 1993. State social intervention and labour regulation: the Argentine case, Cambridge Journal of Economics, vol. 17, no. 4 DeFreitas, G. and Marshall, A. 1995. Labour surplus, worker rights and productivity growth: a comparative analysis of Asia and Latin America (mimeo) de Janvry, A., Graham, A., Sadoulet, E., Espinel, R. and Spurrier, W. 1994. The Political Feasibility of Adjustment in Ecuador and Venezuela, Part 1: Ecuador, Paris, OECD Development Centre Studies Dombois, R. and Pries, L. 1994. Structural change and trends in the evolution of industrial relations in Latin America: a methodological-conceptual outline, XIII World Congress of Sociology (mimeo) Ellner, S. 1993. Organized Labour in Venezuela, Wilmington, Delaware, SR Books Emerson, M. 1988. Regulation or deregulation of the labour market, European Economic Review, vol. 32,775-817 European Commission. 1994. Tableau de Bard, 2, DGV, Brussels Feldman, S. 1995. Contratos temporales de trabajo promovidos: su uso en el sector privado, Estudios del Trabajo, 8/9 48 A. Marshall Hofman, A. A. and Buitelaar, R. M. 1994. Ventajas comparativas extraordinarias y crecimiento a largo plaza: el caso de Ecuador, CEPAL Review, vol. 54 Labour Canada. 1993-94 Employment Standards Legislation in Canada, Ottawa, Labour Canada L6pez Castaiio, H. 1993. Contexto macroecon6mico colombiano, mercado !aboral urbana y retos para una politica de empleo, FESCOL (Fundaci6n Friedrich Ebert de Colombia), Documentos de Trabajo, Bogota Marshall, A. (ed.) 1990. El empleo publico frente a la crisis. Estudios sabre America Latina, Geneva, International Institute for Labour Studies (ILO) Marshall, A. 1992. Circumventing Labour Protection. Non-standard Employment in Argentina and Peru, Research Series no. 88, Geneva, International Institute for Labour Studies Marshall, A. 1994a. Economic consequences of labour protection regimes in Latin America, International Labour Review, vol. 133, no. 1 Marshall, A. 1994b. State intervention, the labour market and inequality in Argentina (1976-1994), in Berry, A. (ed.), Labour Market and Income Distribution Outcomes of Economic Reform in Latin America (in press) Marshall, A. 1996. Employment protection reforms in South America: effective instruments to manage employment? FOCAUCIS Discussion Papers FC1996-1, University of Toronto, Canada Mosley, H. G. 1994. Employment protection and labor force adjustment in EC countries, in Schmidt, G. (ed.), Labor Market Institutions in Europe, Armonk-London, M. E. Sharpe Muckenberger, U. 1989. Non-standard forms of employment in the Federal Republic of Germany: the role and effectiveness of the State, in Rodgers, G. and Rodgers, J. (eds), Precarious Jobs in Labour Market Regulation, Geneva, International Institute for Labour Studies Nairn, M. 1993. Paper Tigers and Minotaurs, The Politics of Venezuela's Economic Reforms, Washington, D.C., The Carnegie Endowment for International Peace Nairn, M. 1995. Latin America the morning after, Foreign Affairs, vol. 74, no. 4 OECD (Organization for Economic Cooperation and Development). 1993. Employment Outlook, July Paus, E. A. 1994. Economic growth through neoliberal restructuring? lnsights from the Chilean experience, Journal of Developing Areas, vol. 28, October Pita, E. 1993. Cambios institucionales en el mercado de trabajo y precarizaci6n del empleo en el Ecuador, IILS/PREALC, Santiago, May (mimeo) Prima Braga, C. A., Nogues, J. and Rajapatirana, S. 1995. Latin America and the Caribbean in the world economy: trade reforms-progress and risks of reversal, Annual Bank Conference on Latin America and the Caribbean, Rio de Janeiro, June (mimeo draft) Rebitzer, J. B. 1987. Unemployment, long-term employment relations, and productivity growth, Review of Economic and Statistics, vol. 64, 627-635 Reyes Posada, A. 1994. Employment problems and challenges in Colombia, OECD International Forum on Latin American Perspectives, Paris (Mimeo) Romaguera, P., Gonza!es, P., Mizala, A. and Montero, C. 1995. Labour flexibility: the case of Chile (mimeo) Seminario, B. and Galarza, E. 1993. Informe sabre la coyuntura econ6mica peruana 1992-IV, Boletin de Opinion, no. 8, Lima, Consorcio de Investigaci6n Econ6mica Verdera, F. 1995. Efectos del ajuste y la reforma !aboral sabre el mercado de trabajo de Lima: 1990-1994 (draft mimeo) Weisskopf, T. E. 1987. The effects of unemployment on labour productivity: an international comparative analysis, International Review of Applied Economics, vol. 1, 127-151 You, J. 1992. Labour institutions and economic development in Korea, International Workshop on Labour Institutions and Economic Development in Asia (mimeo) Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Weakening employment protection in Latin America: incentive to employment creation or to increasing instability? Creator An entity primarily responsible for making the resource Adriana Marshall Source A related resource from which the described resource is derived International Contributions to Labour Studies 6, pages 29-48 Date A point or period of time associated with an event in the lifecycle of the resource 1996