Eugene Kouassi

Health expenditure and growth dynamics in the SADC region: evidence from non-stationary panel data with cross section dependence and unobserved heterogeneity

This paper investigates the long run relationship between health care expenditure and economic growth, using panel data for 14 Southern African Development Community (SADC) member countries over the period 1995–2012. The non-stationarity and cointegration properties between health expenditure per capita and GDP per capita were examined, controlling for cross section dependence and heterogeneity between countries. Our results suggest that health expenditure and GDP per capita are non-stationary and cointegrated. These findings seem to confirm the notion that health expenditure is non-discretionary—health is a necessary good—in the SADC region. The estimated income elasticity is below unity but higher than what was obtained for the OECD regional grouping. The policy implication of our result is that adequate health care service provision should be a key objective of governmental intervention in the SADC region.

Conditional Score Tests for Heteroscedasticity in the Two-Way Error Components Model

This paper extends the one-way heteroskedasticity score test of Holly and Gardiol (2000, In: Krishnakumar, J, Ronchetti, E (Eds.), Panel Data Econometrics: Future Directions, North-Holland, Amsterdam, pp. 199–211) to two conditional Lagrange Multiplier (LM) tests of heteroskedasticity under contiguous alternatives within the two-way error components model framework. In each case, the derivation of Raos efficient score statistics for testing heteroskedasticity is first obtained. Then, based on a specific set of assumptions, the asymptotic distribution of the score under contiguous alternatives is established. Finally, the expression for the score test statistic in the presence of heteroskedasticity and related asymptotic local powers of these score test statistics are derived and discussed.

A Joint Score Test for Heteroscedasticity in the Two Way Error Components Model

This article extends the work by Holly and Gardiol (2000) (A score test for individual heteroscedasticity in a one-way error component model. In: Krishnakumar, J., Ronchetti, E., Eds. Panel Data Econometrics: Future Directions. Elsevier, North-Holland, Amsterdam, pp. 199–211, Ch. 10) to the two-way error components model. It deals exclusively with a joint heteroscedasticity test by first deriving Raos efficient score statistics. Then, based on appropriate set of assumptions, we deduce the asymptotic distribution of the score under contiguous alternatives. Finally, we provide the expression for the score test statistic in the presence of heteroscedasticity and discuss its asymptotic local power.

CLIMATE UNCERTAINTY AND AGRICULTURAL SOIL CONSERVATION INVESTMENT DECISIONS

This paper conceptualizes the soil conservation decisions of farmers when confronted with climate uncertainty. Using a dynamic stochastic optimization model with uncertainty captured by climate variability, soil conservation investment is assessed within the framework of an investment adjustment cost model. The theoretical results reveal that the effects of an uncertain future climate on the optimal path of soil conservation investment depend on how sensitive production is to climate and that input and output prices have level effects on the optimal path of investment. An empirical application to data from Texas shows that depending on the sensitivity of production to climate and the severity of the risk; climate induced uncertainty may have a negative effect, a threshold (U-shaped) effect, or no effect at all on soil conservation investment.

Testing for heteroskedasticity and spatial correlation in a two way random effects model

An extension of the Holly and Gardiol (2000) and Baltagi et al. (2009) papers to the two way context, with heteroskedastic and spatially correlated disturbances is considered. One then derives a joint LM test for homoskedasticity and no spatial correlation. In addition, two conditional LM tests are also derived: for no spatial correlation given heteroskedasticity and for homoskedasticity given spatial correlation respectively. These tests are compared with marginal ones that ignore heteroskedasticity in testing for spatial correlation, or spatial correlation in testing for homoskedasticity. Monte Carlo results show that the LM and LR tests perform well even for small N and T whereas their Wald counterparts tend to oversize. An application on the demand for cigarettes is also conducted. The misleading inference can occur when using marginal rather than joint or conditional LM tests when spatial correlation or heteroskedasticity is present.