Reparameterization of Vector Error Correction Model from Auto-Regressive Distributed Lag to Analyze the Effects of Macroeconomic Shocks on Youth Employment in Kenya

Abstract

This study analyzes the effects of reparameterization of autoregressive distributed lag (ARDL) to vector error correction model (VECM) through cointegration of time series. It further verifies the effects of macroeconomic shocks on youth unemployment in Kenya using VECM. First, the unit root test has been done on youth unemployment (YUN), gross domestic product (GDP), external debt (ED), foreign direct investment (FDI), private investment (PI), youth literacy level (LR), and youth population (POP) to verify stationarity. The Johansen Cointegration Test has been employed and revealed three long run relationships which can be interpreted as a GDP effect, External Debt effect and Foreign Direct Investment effect relations. A structural VECM has been described through restrictions derived from the Cointegration Analysis. Based on the results of the Impulse-Response Function analysis and variance decomposition analysis of the Structural VECM, it is concluded that GDP, literacy level, population, Private Investment, External and FDI shocks have significant effects on Kenyan youth unemployment in the long run. Based on the results of the Impulse-Response Function and variance decomposition analyses of the Structural VECM, it is concluded that GDP, literacy level, population, and FDI shocks have significant effects on Kenyan youth unemployment in the long run. Whereas population, external debt, private investment, and GDP have positive effects, foreign direct investment and literacy rate have negative effects on youth unemployment in the long run. The results provide a statistical basis for assessing and prioritising investment policies and initiatives to maximise youth employment and attain the demographic dividend.