Thursday, May 3, 2012

Review of political connections (by Francesc Trillas)

I have been working on the review of 18 empirical papers on political connections for my research with Pau Castells on this topic in Spain. Previous research either ignored  or drammatically underestimated  the extent of political connections by large firms in Spain.
A perfect comparison of our findings to those for other countries is impossible because no author has used exactly the same definition of political connections. However, keeping in mind the differences in definitions and sample bases, we cautiously claim that the extent of political connections in Spain according to our data (50 out of 69 firms, i.e. 72%, are politically connected) is at the higher end of the presence of political agents in boards of directors. Actually, the percentage is close to the percentage of politically connected privatized firms in developing countries.
-As opposed to most papers (analyzing the US, Germany and several developing countries as well as international cross-country evidence), but similarly to evidence for France, we find a negative correlation between political connections and corporate performance, although in Spain the connections cannot be attributed to an elite of highly trained technocrats as in France.
Possible (not necessarily alternative) interpretations of why connections in Spain have a negative impact on performance (going against the findings of most, but not all, of the literature in many other countries and cross country evidence):
-Agency explanation and weak corporate governance. This is consistent with the view that many large Spanish firms were state-owned firms that were privatized with a dispersed shareholding that were thought to need protection from takeovers and which the government still tried to control to some extent after public ownership. This is consistent with previous work of mine on Telefonica and Spanish firms in the energy sector.
-Political “contributions” by firms as consumption as opposed to investment, as argued by those who think there is too little, not too much, political investment by firms relative to the stakes of policies for them.
-Low corruption or good institutions reducing the value of connections.
-Politicians able to redirect the objectives of firms toward the common good and away from profit maximization. However, the few papers that analyze the social costs of political connections show that there are high net social costs. This suggests that the diversion of objectives is not towards the common good but towards other objectives held by connected politicians.
-In the bargain by which firms relinquish a portion of control rights in exchange for subsidies and protection, politicians have more bargaining power (but this explains a zero correlation between firm profits and connections, not a negative correlation, unless there is some sort of “winner’s curse”).
-Reverse causality: firms with bad performance recruiting politicians to obtain protecting policies.
-A behavioral explanation: firms recruit politicians because they are well known (availability bias) and are thought to be able (after analyzing their political career) to stir circumstances in the direction of the firm’s objective (attribution bias).
We’ll keep working.

No comments:

Post a Comment