Regional Unemployment in Egypt: Spatial Panel Data Analysis

Document Type


Publication Date

Winter 12-2020


Egypt's rate of unemployment averaged 11% over the period 1993–2018. Although this has motivated abundant research, it has focused on nation‐wide unemployment, with little attention to unemployment at the regional (i.e., governorate) level. Indeed, there are wide disparities in unemployment among governorates. Moreover, high unemployment governorates may be clustered together in a way that yields spatial autocorrelation. The present paper therefore fills this gap in empirical literature by analysing unemployment by region using a spatial panel data model (for 27 governorates over the 2006–2016 period). It establishes the existence of (positive) spatial dependence in regional unemployment. Regional unemployment is further found to be affected by low levels of education among the governorates' populations (illiterates, and holders of intermediate and above intermediates) pointing to a mismatch between the skills demanded and the skills supplied at the governorate level. Though services have the highest share in average employment across governorates (0.64), followed by agriculture (0.26), neither of the two indicators (proxies) of activity mix is found to be statistically significant. Other variables expected to affect regional unemployment were not statistically significant (female and youth unemployment, despite high prevalence of both female and youth unemployment across the governorates of Egypt). Policy implications include: the pooling of efforts between neighbouring governorates to address unemployment; matching leading activities in the governorates with the ranks of the unemployed; using apprenticeship schemes; increasing female labour force participation; and promoting private sector job creation.