Multilevel Analysis of Factors Underlying in‐Work Poverty: Evidence From Türkiye

ABSTRACT In‐work poverty is a prevalent phenomenon in many countries and creates significant challenges from various aspects. This study investigates the underlying factors of in‐work poverty utilizing Türkiye as a case study and using the latest available (2022) microdata from the Survey of Income and Living Conditions (SILC). It first measures the at risk of poverty or social exclusion rate (AROPE) among the working population, revealing that 22.13% of workers suffer from poverty and there is a considerable sectoral variation. Subsequently, the underlying demographic characteristics, along with job‐related and regional factors of in‐work poverty, are examined through multilevel logistic regression analysis. Among demographic characteristics, being older and having higher education levels are associated with lower probabilities of in‐work poverty, while household dependency raises this risk. Analyzing the association between in‐work poverty and job characteristics indicates that public sector employment and social security registration lower the probability of in‐work poverty. Furthermore, compared to the self‐employed, employers are much less likely to be poor, while employees are more vulnerable. Finally, living in a region with a higher GDP per capita lowers the probability of working poverty, whereas regions with higher Gini indices, unemployment rates, and Syrian refugee rates are associated with higher poverty risks.