This question began to attract attention in the 1990’s. Mobarak’s (2005) findings in the Review of Economics and Statistics provide a good marker on the current conventional wisdom on one aspect of the question under consideration: Democracy has no direct effect on average growth but it does reduce the volatility of growth and the latter impacts growth negatively. The logic underlying the result makes sense as checks and balances reduce the probability of economically very attractive or unattractive projects being undertaken by a society. The result implies that democracy has no effect on the average quality of economic projects undertaken by a democratic society relative to an authoritarian one but it reduces the range symmetrically.
A separate strand of literature has looked at the relationship between democratization and growth. Here the argument is that while levels of democracy may not be related to growth directly, changes from authoritarianism to democracy seem to increase growth while changes from democracy to authoritarianism have the opposite effect. These findings are reported, for example, by Pappaionnau and Siouronnis (2008) in the Economic Journal. They also show how their findings coincide with those of other author using different measures. The logic of their argument also makes sense in that it relies on the belief that the benefits of democracy appear in the long run and other authors who disagree may not have looked for them from that perspective.
Finally, another strand of literature finds that liberal democracy in terms of one component of civil liberties has a robust positive impact on economic development. BenYishay and Betancourt (2010) take advantage of Freedom House’s decision to disaggregate their political rights and civil liberties indexes, respectively, in 2006. Their findings, reported in the Journal of Institutional Economics, show that (when you measure democracy in terms of FH political rights index or civil liberties index) neither index has an effect on economic development measured in terms of per capita income. Nonetheless, when you use the disaggregated components, one and only one of the seven components of the two indexes has a positive and robust impact on economic development. This component is G, an index of personal autonomy and individual rights, and one of the components of the aggregate civil liberties index. We pursue the implications of this result below and in subsequent postings. The article is available by clicking here.
The components of the political rights index are: A. Electoral Process, which consists of three questions on the freedom and fairness of elections for the head of state and the legislature; B. Political Pluralism and Participation, which consists of four questions on political parties, their ability to get elected ( even if they are minorities) and their subservience to powerful groups; C. Functioning of Government, which consists of three questions on the ability of elected officials to determine policies, be free from pervasive corruption, and accountable in transparent ways.
The components of the civil liberties index are: D. Freedom of Expression and Belief, which consists of four questions on this topic; E. Freedom of Assembly, which consists of three questions on this topic; F. Rule of Law, which consists of four questions related to this topic; and G. Personal Autonomy and Individual Rights, which consists of four questions on mobility rights, security of property rights, social freedoms, and citizens’ tolerance in the economic sphere. Mobility rights assure assignment of resources to highest productivity activities; security of property rights lowers uncertainty and transaction costs and is indispensable for operation at a high level of transactions in modern markets.
[D. Acemoglu, democracy, economic development, political rights, civil liberties, institutions, J. Robinson, political economy, dictatorship]