Survival Analysis of Regional Unemployment in Uganda: Evidence from the Uganda National Panel Survey (UNPS)

Abstract

The study utilizes survival analysis methodology to examine unemployment duration in Uganda's five regions. The analysis is developed on a database of cohorts of the Uganda National Household Survey (UNHS) for the year 2005/06 and the Uganda National Panel Survey (UNPS) for the years 2009/10, 2010/11 and 2011/12. This work analyses individuals' first spell of unemployment using the Kaplan Meier estimator of the survival functions in unemployment and the covariate effects on duration using the Cox Proportional Hazard Model. Results suggest that over 12 months cohorts in the Eastern and Northern regions face longer spells of unemployment with a 0.85 probability of unemployment. However, in the long run and with the exception of Western Uganda, unemployment duration in all regions is not statistically different from that of Kampala. Cohorts with higher education gain employment slower than those with less education across all regions. Women exit unemployment much slower than men if the unemployment duration is less than 12 months and much faster where the duration is longer than a year. Exit from unemployment increases with age. Investments to curb unemployment should focus on equipping youth younger than 25 years with employable skills.