Sami Chaabouni

The Determinants of Health Expenditures in Tunisia: An ARDL Bounds Testing Approach

This article examines the determinants of health expenditures in Tunisia during the period 1961-2008, using the Autoregressive Distributed Lag (ARDL) approach by Pesaran et al. (2001). The results of the bounds test show that there is a stable long-run relationship between per capita health expenditure, GDP, population ageing, medical density and environmental quality. In fact, on the one hand there are the short-run and long-run results which reveal that health care is a necessity, not a luxury good. On the other hand, results of the causality test show that there is a bidirectional causal flow from health expenditures to income, both in the short and in the long run.

The dynamic links between carbon dioxide (CO2) emissions, health spending and GDP growth: A case study for 51 countries

Abstract This document investigated the causal relationship between carbon dioxide (CO2) emissions, health spending and GDP growth for 51 countries (divided into three groups of countries: low‐income countries; group of countries with lower and upper middle income; group of middle income countries) covering the annual period 1995–2013. Dynamic simultaneous‐equations models and generalized method of moments (GMM) are used to investigate this relationship. The main results provide evidence of a causal relationship between the three variables. The empirical results show that there is a bidirectional causality between CO2 emissions and GDP per capita, between health spending and economic growth for the three groups of estimates. The results also indicate that there is a unidirectional causality from CO2 emissions to health spending, except low income group countries. We found that health plays an important role in GDP per capita; it limits its effect on a growing deterioration in the quality of the environment. HighlightsRelationships among carbon dioxide (CO2) emissions, health spending and GDP growth are examined.Bidirectional causality between CO2 emissions and GDP per capita for three groups of countries.Bidirectional causality between health spending and GDP per capita for three groups of countries.Unidirectional causality from CO2 emissions to health spending, except low income group countries.

Efficiency of public hospitals in Tunisia: a DEA with bootstrap application

This paper analyses the technical efficiency of public hospitals in Tunisia during the period 2000–2007. We use a two-stage procedure proposed by Simar and Wilson (2007). In the first stage, we use data envelopment analysis (DEA)-VRS model to identify the efficiency scores, and the second stage, we use a bootstrapped truncated regression model to identify the covariates that explain technical efficiency. We confirm that the efficiency growth did not improve significantly over the period of this study. In addition, we show that the efficiency of public hospitals is homogenous, with little variation across the analysed hospitals.

Productivity and its components in Tunisian public hospitals

This paper analyses the productivity and efficiency of the Tunisian public hospitals over the period from 2000 to 2007. We obtain estimates of productivity using the bootstrapped Malmquist index, and a truncated regression model is used to examine how external factors impact the level of productivity and its components (i.e. (a) the bed occupancy ratio which represents the hospital capacity utilisation; (b) the average length of stay; (c) the total number of beds). The hospitals are represented by means of a simple model whereby they use inputs (labour and capital) to produce outputs (healthcare services). The efficiency model, which is output–oriented, assumes variable returns to scale. The results show that the evolution of the productivity of the hospital is overwhelmingly due to the technological change, but the magnitude of the change in technical efficiency remains low. The results suggest also that the average length of stay has a negative impact on productivity and its components whereas other variables like, the bed occupancy ratio and the total number of beds, are non–significant.