The African Business Review May-Jun 2014 | Page 30

The aim of our model is to show that democracy affects negatively inflation and to evaluate the explanatory weight of democracy in the presence of proxy variables of economic policies. Variables (MONgrowth) and (InRES) represent these proxy. Other control variables are used. It comes to annual percentage growth rate of GDP[11] (Growth), with the assertion that the increase in GDP growth rate, which corresponds to an increase in productivity, will lead to price decrease, and imports plus exports as a percentage of GDP[12] (Trade) to control for trade openness. These two variables were used as control variables by Desai, Ologsfard and Yousef (2002) in a study similar to ours, but the goal was to demonstrate that the impact of democracy on inflation remains conditioned by the degree of income inequality.
 Empirical strategy the period 1996-2012. The sample of countries covering this first estimation is heterogeneous, containing developed, developing and poor countries. The economic relationship we are interested in identifying is:
 INFit = α + β1 (DEM)it + β2 (POLSTAB)it + β3 (MONgrowth) it + β4 (LnRES)it + β5 (Growth)it + β6 (Trade)it + εit (1) Where (εi) is the random error term.
In a first step we have run a Haussman specification test on whether fixed or random effects are more convenient to our data’s. The Haussman statistic is: H= (β^FE - β^RE)’ [ Var (β^FE) - Var (β^RE)] (β^FE - β^RE) ~ 2x(k) The first estimation we carried out covers 124 countries and for The Haussman specification test is based on the following corps 11 12 Annual percentage growth rate of GDP at market prices based on constant local currency. Aggregates are based on constant 2005 U.S. dollars. Source: World Bank national accounts data, and OECD National Accounts data files. 30 | The African Business Review The sum of exports and imports of goods and services measured as a share of gross domestic product. Source: World Bank national accounts data, and OECD National Accounts data files.