Theses Doctoral

Essays on the Political Economy of International Agreements

Lazarevski, Goran

This dissertation consists of three essays that sit at the intersection of international trade, political economy and the economics of innovation. It analyzes from a critical perspective the relationship between organized interest groups and international agreements on trade and intellectual property rights (IPR) protection and offers new theoretical insights, which it then supports empirically.
My first essay calls into question the logic of the standard Grossman-Helpman/Bagwell-Staiger model of trade agreements, according to which governments enter international treaties to prevent terms-of-trade manipulation and special interest politics has a trivial role. Despite its immense popularity, it remains inconsistent with observed trade policy and with the practitioners' understanding of trade treaties. By assuming that subsidies have additional political cost beyond their monetary cost, I show how international agreements result in the reduction of political protectionism through the crucial role of exporting lobbies in the negotiations process. At the same time, the model resolves three prominent puzzles in the literature: the terms-of-trade puzzle, the anti-trade bias puzzle and the inefficient redistribution puzzle. Finally I find empirical support for the model and my key assumption using data on US agricultural trade policy.
In the second essay I propose a model that considers the effect of firm lobbying for IPR protection in an international setting in innovation-driven economies. In particular, I compare the IPR protection level and global social welfare between the case when countries set their IPR policies non-cooperatively and when they enter an international treaty, such as the TRIPS, TPP and TTIP. I find that lobbying necessarily leads to inefficient international agreements resulting in too much IPR protection and may even be welfare-reducing relative to no cooperation. I also show that international lobbying and high concentration of capital can further exacerbate this outcome. The model generates predictions consistent with patterns I find in the data on US firms' lobbying expenditures and the value of their international patent portfolios.
Finally, the third essay provides a critique of a popular structural patent valuation methodology that utilizes the stock market response to news about patent grants, first introduced by Kogan et al. (2012). Using their methodology (refined and improved in terms of the theoretical derivation), I perform a placebo estimation of US patent values and compare the results with the true patent value estimates as per Kogan et al's paper. I find strong evidence that the "true" patent value estimates are not driven by patent news announcements, but rather are an artifact of the estimation methodology itself and as such cannot be used for comparisons across different patent-holding firms and grant years. I further corroborate the external validity of this critique by applying the same method to a novel database of Chinese patents and finding that the same conclusion holds.


  • thumnail for Lazarevski_columbia_0054D_14418.pdf Lazarevski_columbia_0054D_14418.pdf application/pdf 1.1 MB Download File

More About This Work

Academic Units
Thesis Advisors
Naidu, Suresh
Ph.D., Columbia University
Published Here
February 2, 2018