The Politics of Market Selection

This work seeks to determine the effect of political power of economic sectors in the policies their plants face, in their market survival probabilities, and in the aggregate productivity of the economy. Data from colombian manufacturing sector at the plant level for the period 1990-1998 is used to estimate several empirical models. Also, political power variables are constructed: dummy variables that indicate membership to economic associations and economic corporate groups, and an index that captures both the economic importance of the plant´s economic sector in the geographical region where it is located, and the electoral influence of that region in the national political scene. In the first place, we estimate the effect of political power on policy variables. For these, we use nominal tariffs, mainly because they show differential treatments between sectors, because they affect all sectors in the economy, and because they represent policies in a quantitatively way, and thus are subject to quantitative modeling. We found that political power of sectors has a significant and positive effect on the tariffs their plants face. This happens both when it comes from belonging to an organized group that can perform lobby activities and when it comes from the political support that the sector´s welfare implies for policy makers. In the second place, a market selection model is estimated to determine the effect of political power on the probability of exit of a plant. We found that a plant that belongs to a more politically influential sector can survive in the market with productivity levels that would otherwise make it go out of business (compared to plants belonging to sectors with less political power). Third, the economy´s aggregate productivity is calculated with and without the effect of political power. We find that, since this effect allows less productive plants to survive, aggregate productivity is decreased an average of 9.86% per year.