Work histories and the access to contributory pensions: the case of Uruguay

Abstract

ResumenHistorias laborales incompletas y altamente fragmentadas amenazan con dejar a muchos contribuyentes de los regímenes de pensiones de América Latina sin la pensión mínima garantizada, o incluso sin acceso a la pensión (jubilación) común. En el presente estudio, se propone una metodología para evaluar este riesgo, identificar los grupos vulnerables y estudiar los posibles factores determinantes de las historias de contribuciones, utilizando la información de los registros de historia de laboral de las instituciones de seguridad social. Aplicando esta metodología sobre los registros de la principal institución de seguridad social del Uruguay, el Banco de Previsión Social, se obtiene que la mayoría de los contribuyentes de esta institución podría no cumplir con el mínimo de años de cotización requeridos actualmente para acceder a una pensión (jubilación) común al llegar a las edades habituales de retiro.Palabras clave: densidad de cotizaciones, historia laboral. Clasificación JEL: H55, J14, J26. AbstractIncomplete and highly fragmented work histories threaten to leave many contributors of the pension schemes in Latin America without the minimum pension guarantee or even without access to the ordinary pension. We propose a methodology to assess this risk, identify vulnerable groups and study potential determinants of the history of contributions using information from the work history records of the social security institutions. We apply this methodology to the largest social security institution of Uruguay, the Banco de Previsión Social, and show that the majority of contributors to this institution might not comply with the minimum number of years of contribution that is currently required to access an ordinary pension when they reach the retirement age.

Extending access to contributory pensions: The case of Uruguay

“…The probability of contributing is strongly dependent on prior contributory history, so it is relevant to consider such dependence. Although conditioning only on the immediate previous month may seem insufficient, Bucheli, Forteza and Rossi (), who conditioned on the entire observed history, found qualitative results that were no different to those of Bucheli et al () and Bucheli, Forteza and Rossi (). In the second stage, we use the probabilities estimated in the first stage to simulate complete work histories.…”

“…Bucheli, Forteza and Rossi () used the same sample of work history records and pursued similar objectives, but they conditioned the probabilities of contributing on all preceding information. In the first stage, they estimated transition probabilities between contributing and not contributing using survival analysis.…”

A model for the pension system in Mexico: diagnosis and recommendations

“…Alonso (2003) showed that in the Spanish case, education was a very relevant variable for including transition dynamics in which the changing composition of the level of studies achieved by a population could vary the projections for the pension system significantly. This has been a trend in recent analysis of pension systems in emerging countries (Berstein et al, 2005;Bernal et al, 2008;Bucheli et al, 2010;Forteza et al, 2011). In our analysis we conclude the same, as the potential salary profiles observed for men and women and by age groups are notably different, as seen in the following sections, depending on the difference between the levels of studies achieved.…”

“…The model we propose for the IMSS fits into this second group; this methodology has been used in recent literature for instance, Berstein et al (2005), Bernal et al (2008), Bucheli et al (2010) and Forteza et al (2011). The richness of the analysis increases as various sensitivity exercises are carried out.…”

Social Security and Retirement in Uruguay

“…A sample of the work history records was made available to the research community and have been used to analyse the impact of social security reforms on individual behaviour and the macroeconomy (Álvarez et al, 2010Bucheli, Forteza, & Rossi, 2010;Forteza et al, 2009). Existing studies are based on the estimation of reduced form models, though, and are not immune to the Lucas critique: the policy response parameters being estimated are not necessarily invariant to policy changes.…”