Haïti à l’épreuve du covid-19 : impact économique potentiel du virus et détermination du seuil optimal de dépenses publiques qui maximise la croissance économique du pays

The main purpose of this paper is twofold (i) to estimate by using a neoclassical model of growth proposed by IMF, the economic and financial impact of COVID-19 in Haïti; (ii) to analyse threshold effects of government size on economic growth in this country, by using Hansen’s Threshold Auto Regress...

Full description

Bibliographic Details
Main Author: Louis Dupont
Format: Article
Language:English
Published: Université des Antilles
Series:Études Caribéennes
Subjects:
Online Access:https://journals.openedition.org/etudescaribeennes/21783
Description
Summary:The main purpose of this paper is twofold (i) to estimate by using a neoclassical model of growth proposed by IMF, the economic and financial impact of COVID-19 in Haïti; (ii) to analyse threshold effects of government size on economic growth in this country, by using Hansen’s Threshold Auto Regressive (TAR), 1999. It uses the time series data over the period 1980-2019. In the first model, the paper finds that on account of the pandemic, the Haïti’s GDP could be reduced of 14.1% in 2020. In the second model, our results provide strong evidence for the existence of a non-linear relationship between government expenditure and economic growth. Based on these results, the paper finds that the optimal level of government expenditure that maximizes economic growth for Haïti is 14.1% of GDP against a staff level of 11.1%. They show that the state spends well below its potential growth. On the other hand, if this threshold is exceeded, the increase of the percentage ratio of state budget expenditure/GDP will have a negative impact on the speed of GDP growth. Thus, our recommendations are in favor of increasing government investment to education, health, infrastructure, environmental sectors … that are neglected up to this day in Haïti. However, these expenditures have to be limited to a level that no excess the optimal size
ISSN:1779-0980
1961-859X