~5~
International Journal of Statistics and Applied Mathematics 2019; 4(1): 05-17
ISSN: 2456-1452
Maths 2019; 4(1): 05-17
© 2019 Stats & Maths
www.mathsjournal.com
Received: 04-11-2018
Accepted: 08-12-2018
Shem Otoi Sam
PhD Student, School of
Mathematics, University of
Nairobi, Kenya
Ganesh P Pokhariyal
Professor, School of
Mathematics, University of
Nairobi, Kenya
MM Manene
Professor, School of Mathematics
University of Nairobi, Kenya
Dr. Isaac C Kipchumba
School of Mathematics
University of Nairobi, Kenya
Correspondence
Shem Otoi Sam
PhD Student, School of
Mathematics, University of
Nairobi, Kenya
Reparameterization of vector error correction model
from auto-regressive distributed lag to analyze the
effects of macroeconomic shocks on youth employment
in Kenya
Shem Otoi Sam, Ganesh P Pokhariyal, MM Manene and Isaac C Kipchumba
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.
Keywords: structural error correction model, cointegration
1. Introduction
The demographic dividend, defined as the opportunity to achieve rapid socio-economic
development resulting from decline in fertility levels and targeted investments in intensive
employment sectors, has been deemed a solution to many problems being experienced by
African countries. The African Union adopted a common position on the Post 2015
Development Agenda that resulted in the incorporating demographic dividend in the 2030
Agenda for Sustainable Development. In this way, the African Union formulated the theme
“Harnessing the Demographic Dividend through investments in Youth” for the year 2017.
The Kenya constitution defines youth as persons between the ages of 18 and 34. In Kenya, the
youth constitute 35% of the population. The youth in Kenya are experiencing much higher
unemployment rates 67% compared to the rest of population at 34%.
The Kenyan labour market is one that is characterized by inadequate employment
opportunities against a large and growing population of unemployed young people. It is dual in
nature, presenting a small formal sector alongside a large informal sector. Over 30% of those
on wage employment are casuals. Youth with primary education are in formal employment
4%, informal employment 54%, students 14% and unemployed 28%.
Those with secondary education in: formal employment 12%; informal employment 40%;
students 26%; and unemployed 22%. While those with tertiary education are in: formal
employment 31%;