In Latin America, the debate on what constitutes a ‘good’ or ‘bad’ job has been dominated by the phenomenon of informality. Indicators like the ‘informal sector size’ or the proportion of workers in ‘informal employment’ give little attention to the intrinsic features of jobs that affect workers’ well-being, thus misleading policy efforts. Validation of alternative and comparable human-centred measures of job quality (JQ) is needed.

This study aims to evaluate the validity of a multi-dimensional measure of JQ in developing countries, and its usefulness against narrow indicators of formality/informality. To this end, Sen’s capability-approach is used along with Green and Mostafa’s operationalisation of JQ (Eurofound, 2012), which considers dimensions as varied as earnings, career prospects, autonomy, intensity, social environment, physical environment, and working time.

With Central America as the research setting, I address four questions: (1) does Eurofound’s indicator capture JQ inequalities at the individual level? (2) Can we draw meaningful comparisons between countries about their ability to provide good jobs? (3) Are the selected features of what constitutes a good job positively associated with Central American workers’ well-being? (4) Is the concept of JQ attuned with what local experts conceive as a ‘good job’? The research uses a mixed-methods approach to analyse the First Central American Survey on Working Conditions and Health – conducted in 2011 in Panama, Costa Rica, Nicaragua, Honduras, El Salvador and Guatemala – in addition to semi-structured interviews with selected informants from these six countries.

The results obtained show, firstly, a reasonable distribution of JQ across groups of workers. They confirm that formal jobs are not ubiquitously the best quality jobs. Secondly, the results evidenced significant variation at the country level regarding earnings and intrinsic job quality, with Costa Rica often ranking at the top. Interestingly, JQ rankings do not always follow from countries’ industrial structure, economic performance, informal sector size, or other developmental indicators of common usage; country differences in JQ appear associated with the practical enforcement capacity of labour institutions like trade unions, inspection systems, and the state itself. Thirdly, I corroborated that the selected job features have a positive impact on Central American workers’ well-being (except, puzzlingly, for work-time related aspects). Moreover, the positive health effect associated with performing in an intrinsically good job proved to be greater than the effect of working formally. Lastly, I confirmed that local perspectives about what constitutes a ‘good job’ are in great part consistent with the features included in Green and Mostafa’s JQ scheme, while other intrinsic dimensions of the framework have struggled to enter the public discourse.

These findings indicate that a JQ framework is generally valid in the Central American context, and provides more information than a conventional indicator of informality. The study contributes to extend the capability approach to the realm of work and to stress its potential for international comparative research. It is recommended that countries collect richer data about those aspects of jobs that have been proven to affect workers’ well-being significantly and are not revealed in unidimensional informality figures.