This study found a clear and strong association between informal employment and poor self-perceived health in both men and women and in most of the LAC countries. This association is consistent with previous studies in the Latin America region. The main mechanisms that could explain this relationship include precarious employment, and working and living conditions involving income insecurity, poverty, absence of social protection benefits, and lack of access to health services, which characterise informal employment..
However, this association was broader and more significant in countries with statalist welfare state regimes than in familialist regimes. In this sense, the welfare state, as a complex system of regulation and wealth redistribution, seems to partially buffer this effect on population health, as informality and poor SPH were weaker in familialist countries than in statalist countries, where informality and poor SPH are somewhat lower and the welfare states are more developed. This result seems to contradict other studies, which show that more egalitarian welfare states tend to reduce health inequalities. Yet, none of these studies compared the health of informal and formal workers.
The lower prevalence of poor SPH in the working population of statalist countries is most likely the result of a greater proportion of legally and socially protected workers in the formal labour market. In this regard, formal workers could have access to social and health care services, while informal workers remain unprotected. This greater health disparity between formal and informal workers in the statalist countries could be a consequence of the positive impact that labour policies, social protection and health care have on the formal workers, but it fails to improve the living and working conditions of the large proportion of informal workers and their access to health systems and work-related social security. It seems necessary to set labour policies that consider informal employment, linking the labour market with the public health policies to not only reduce the impact of informal employment, but to promote formal jobs an avoid poor working conditions.
On the other hand, the small health difference between formal and informal workers in familialist countries, especially among women, could be explained by two intertwined hypotheses. First, the weak role of the state in familialist countries. In general, familialist countries have lower gross national income per capita and low social and health expenditures. Consequently, social and health services are poor and, in most of the cases, do not even cover the health needs of the formal workers. In this regard, access to proper health services is mostly determined by the individual income level. Most of the formal workers cannot pay for these services and they are in similar circumstances as informal workers without social and health coverages. Second, the global expansion of labour precariousness and deterioration of working conditions have affected all kinds of employment, regardless of whether they have a contract. This generalised precariousness could explain the lack of association between type of employment and health in familialist countries. As other studies have shown, the impact of specific policies, such as unemployment benefits, seems to be associated with worker health in different types of welfare state. Although labour policies are part of welfare state policies and activities, the classification is broader and includes other types of social policies, it would be interesting in future studies to characterise countries according to labour policies based on employment indicators, in order to estimate the impact of labour policies on the association between informal employment and health.
Regarding gender, women reported a higher prevalence of poor SPH than men in the overwhelming majority of countries in the region, which is consistent with other studies. However, the proportion of women and men with informal employment is similar in almost all countries. Previous studies have found a higher prevalence of informal profiles among women compared to men. We also found that women are not more vulnerable than men to the effects of informality, which was shown in previous studies and reflected in more non-significant associations in familialist countries. In addition to the above considerations, these results could be explained by the social security system being less effective in compensating women than men, because women who have formal employment could continue to experience a precarious situation that negatively impacts their health. In addition, a greater proportion of women in familialist countries could be affected by the “double presence” that involves both the productive and reproductive spheres , regardless of the type of employment. This hypothesis will have to be confirmed in future gender-sensitive research, which should include variables that characterise the reproductive sphere and the socioeconomic situation of the family..
This study has some limitations, mainly related to the data sources used. We used the most recent national surveys on working conditions or health surveys available in each country. However, the surveys were not uniform and could slightly differ in the scale used to collect the SPH data. Therefore, comparisons between countries must be made with caution. Furthermore, due to the complexity of measuring and operationalising informality, we used the variable that captures whether the worker has coverage or contributes to a public health/pension plan/insurance to calculate the prevalence of this employment situation. This prevalence represents a proxy of informal employment in the country and could under- or over-estimate the real situation. However, our results for informal employment rates are similar to those reported by the World Bank and those estimated by the International Labour Organisation. Furthermore, this is the most reliable data for measuring the health status of informal workers that the region can currently offer. Official surveys should include variables to better characterise informal employment and its association with health.
In addition, the agricultural working population was excluded from the analysis. Most of the workers in this sector are in an informal situation and likely exposed to worse working and employment conditions than workers in other economic sectors. However, excluding this working population is common practice in most of the studies and a recommendation of the ILO for measuring informality. Moreover, not all LAC countries were included in the study because updated and/or reliable data were not found. However, the analysed sample represents most of the working population in LAC countries. Finally, as in any other cross-sectional study, reverse causality bias could be present, and people with poor health may be more likely to have difficulty finding formal work.
Finally, informal employment has usually been addressed from the economic theory , as it is embedded in a given economic context, and is correlated with other factors (such as labour and social protection, health systems, labour market or demographic evolution) that we were not able to include in this paper. However, in our study, welfare state regimes are used to understand how macro-level determinants influence in health, reflecting economic, social structure and power relations within a society. In this regard, welfare state typology is used as an ecological variable to stratify and assess the possible independent effect of all these factors on the association between informality and health.
This study has several strengths. As far as we know, this is the first study in which the patterns of association between informal employment and health were analysed by gender and welfare state in a large representative sample of LAC countries. Second, the results are based on the most updated high-quality data from a representative sample of LAC countries. Data were from national surveys conducted by official institutions in a representative sample of workers in each country, and data collection was via face-to-face interviews at the workers’ homes. Third, the treatment of the data by a meta-analysis allowed us to give a joint measure of association for the countries according to welfare state regimes and sex. It also allowed us to verify that the heterogeneity within each of these regimes is quite high. Notably, the conceptualisation of welfare states implies ideal types, leaving aside the fact that welfare provision varies greatly between countries of the same regime type. In addition, the typology of welfare regime used in this paper was based on a 2008 analysis of economic and social indicators of labour commodification and welfare decommodification. These indicators have likely changed in some countries over the past 12 years. Therefore, more studies are needed to better understand this complex relationship between the health of workers and the policies of the social welfare regime.