WHEN THE NEW GOVERNMENTS OF UNDERDEVELOPED AREAS supplanted colonial rule, many were optimistic that these governments could operate within a democratic framework. But these nascent governments drifted into authoritarian forms of political life as they faced the exigencies of self-rule. Representative institutions on the Western pattern broke down and were replaced by single-party arrangements headed by charismatic leaders. This form of government, in turn, has tended to give way to bureaucratic and military regimes.
This article presents a new dataset of indicators of political freedom, property rights and political instability for Zimbabwe for the period 1946 to 2005. The dataset is constructed by systematically coding the three concepts of political freedom, property rights and political instability along a multitude of dimensions. The lengthy time coverage of the dataset allows country-specific econometric analysis to evaluate generalizing propositions about the effects of political institutions on economic outcomes. The dataset also contributes to the dynamic analysis of the effects of political institutions on conflict, a contentious issue in political science. Correlations between the new measures reveal that while political instability has a strong and negative relationship with property rights, it has no significant relationship with political freedom. The finding supports the notion that political conflict is significantly higher in the in-between category of semi-democracy than at either end of the democracy scale. The validity of the dataset is supported by its strong correlations with other conceptually and operationally different measures of political institutions. The new dataset has begun to be employed in country-specific time-series studies of the link between institutions and economic outcomes. Two core results are that property rights influence the volumes of foreign direct investment (FDI) in Zimbabwe and that negative spillover effects of poor institutional environments can occur between neighbouring countries. It is feasible to extend the geographical coverage of the dataset by applying our methodological framework to other countries.
