Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/339017 
Year of Publication: 
2025
Series/Report no.: 
Development Research Working Paper Series No. 9/2025
Publisher: 
Institute for Advanced Development Studies (INESAD), La Paz
Abstract: 
This paper uses a multidimensional framework that incorporates constraints of accessibility, affordability, and acceptability to examine the determinants of effective access to the contributory component of Bolivia's pension system. Using nationally representative household surveys from 2005 and 2019, we evaluate the influence of labor-market segmentation, financial capacity, informational barriers, and sociocultural factors on workers' likelihood of affiliation. To address key empirical challenges, including nonlinearity, non-random selection into employment, and perfect or near-perfect prediction, we estimate Probit, Heckprobit, and Firth-Logit models and compute gender-specific average marginal effects. The results indicate persistent structural barriers across periods and settings. Self-employment, unpaid work, and low or unstable earnings consistently reduce affiliation. Informational constraints and distrust were decisive in 2005, while digital access became a critical determinant by 2019. Sociocultural factors, particularly Indigenous identity, also emerged as significant acceptability constraints in the later period. Gender differences in affiliation mainly reflect disparities in employment status, income, and access to information, rather than heterogeneous behavioral responses. Overall, the findings underscore the need for integrated policies that address informational gaps, financial constraints, and labor-market segmentation to bolster access to the contributory component of Bolivia's pension system.
Subjects: 
Pension systems
social protection
informality
labor markets
gender gaps
access barriers
selection models
nonlinear probability models
JEL: 
J26
H55
J21
C35
D14
O54
I38
J16
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.