Explaining non-compliance with labour legislation in Latin America: a cross-country analysis

ADRIANA MARSHALL (GOLDSCHVARTZ)

IILS Discussion Papers

International Institute for Labour Studies, ILO

Lugar: Ginebra; Año: 2007 vol. nc p. 1 - 28

Debates around the effects of labour protective regulations not only deal with their possible impacts on economic, productivity and employment performance, but also include competing views on how the very effectiveness of legislation might be affected. In the literature focussing on Latin America particular emphasis has been placed on the argument that high labour costs and constraints on employers, derived from labour protective regulations, promote non-compliance. Non-compliance has increasingly become an object of concern, yet empirical research on its determinants is scant. In this context, the objective of this paper is to analyse the factors influencing the degree of compliance with labour and social security legislation through a cross-country analysis, in particular the possible explanatory role played by labour regulations themselves, but also by cultural norms, economic strategies, and enforcement levels. The stringency of regulations and levels of non-wage labour costs, if they play a role at all, are but two of the factors that may affect compliance, and their influence may be contingent on other determinants. The term ‘compliance’ refers to employer behaviour vis-à-vis labour and social security legislation (labour codes, collective labour laws, legally binding collective agreements, minimum wages, social security legislation, regulations on health and safety at the workplace), i.e. to whether employers conform with what is stipulated by national legislation in relation to wages, employment and working conditions, and collective labour rights. The article is organised as follows. The analytical background is presented first (section 2). Next, the evolution of compliance levels in Latin American countries from the early 1990s is described, and some preliminary ideas on whether they might be linked to the nature of labour protective regulations or their changes (section 3), and to the pressure of external competition (section 4) are presented. The scope of the effective application of labour legislation in this region by the mid 2000s is examined in section 5. In sections 6 and 7  the indicators for compliance, and the explanatory variables to be included in the model are discussed. Results from exploratory multivariate regression analyses are presented in section 8. The main conclusions are: 1)  Descriptive examination suggests that compliance trends did not follow the pattern often associated in the literature with changes in labour regulation, for instance countries where protection and payroll taxes were reduced nonetheless show increasing non-compliance. Cross-country regression analysis reveals that, of those considered, only one dimension of labour protection, namely costs of dismissal, might have some influence, possibly weak, on compliance levels (the association being negative as usually expected), but that this would be confined to the sector composed of larger private firms and government activities where, actually, non-compliance is much less widespread. Non-wage labour costs do not appear to affect compliance. 2) Non-compliance is a very generalised problem, present in economic sectors producing non-tradeables as well as in those whose output is tradeable. Not only this: non-compliance is much more important in the former, which is not or little affected by international competition. In addition, indicators of external pressure included in the regression models are not associated with degrees of compliance (cross-country analysis), neither do they affect the impact of other explanatory variables. Besides, changes over time in the volume of imports, indicating changes in the degree of competition in the domestic market, do not seem to be correlated with the evolution of compliance. 3) The perception that regulations are easy to evade, i.e. prevailing cultural norms with respect to corruption, which indirectly indicates effective enforcement levels, seems to play an important role in stimulating non-compliance in the smallest firms, precisely those where evasion is localised. This association reflects also how important actual evasion is in determining perceptions of corruption, but the influences act both ways, one reinforcing the other. 4) The effective coverage of labour protective legislation is not as large as would be desirable. Nonetheless, this does not suffice to claim that labour protection should be dismantled. As the descriptive analysis reveals, effective coverage is by no means negligible. Moreover, non-compliance is disproportionally localised in businesses with not more than five employees, and in this micro business sector evasion is fostered by lax enforcement and widespread views of easy corruption, and not by the stringency of labour protective regulations. Therefore, the relaxation of restrictive regulations would have little, if any, impact.