Logo image
International cartels and international trade: theory and evidence
Dissertation   Open access

International cartels and international trade: theory and evidence

Delina Emilova Agnosteva
Doctor of Philosophy (Ph.D.), Drexel University
Mar 2017
DOI:
https://doi.org/10.17918/etd-7302
pdf
Agnosteva_Delina_20171.68 MBDownloadView

Abstract

Cartels Economics
My dissertation studies the links between international cartels and international trade both theoretically and empirically. In the first section of my dissertation, I examine the implications of collusion for trade and welfare in the context of two different theoretical frameworks. First, I build a quantity-setting duopoly model of multi-product firms which interact repeatedly in national markets separated by trade costs to study the impact of collusion on trade and welfare. Each firm produces two goods but has a competitive cost advantage in one. In this setting, an international cartel can choose to shut down production of the inefficiently produced good and import it instead, and thus, can promote trade relative to competition. Further, the cartel extracts surplus by exploiting its market power, but also generates efficiency gains by rationalizing production and trade. Therefore, maximal collusion can welfare-dominate Cournot competition regardless of whether trade costs take the form of transportation costs or import tariffs. Second, I construct a multi-market duopoly model to study the consequences of economic integration for collusive discipline, optimal shipments, and welfare. Firms interact repeatedly in quantities in each other's home markets as well as in third-country markets. When the no-deviation constraint is active, national markets become strategically linked and thus internal and external trade liberalization affects output deliveries and welfare levels in all countries. I characterize the dependence of collusive stability and cartel discipline on trade costs and relative market size. I derive novel results regarding the impact of economic integration on national welfare. For instance, the analysis shows that regional trade liberalization can hurt all countries and the absence of internal trade might be welfare-superior to free internal trade to all nations. In the second chapter, I describe the novel data on international cartels that I have hand-collected. The cartel data cover 173 international cartels and include information on the exact duration of the cartels, the countries of nationality of the cartel-members, as well as the 6-digit product code of the goods subject to collusive activities. Moreover, the data contain various cartel characteristics pertaining to the instruments of collusion (i.e., price-fixing, bid-rigging, sales quotas) as well as details on the practices adopted by the cartel members and the scope of collusion (i.e., cartel's market share). I merge the cartel dataset with the most disaggregated trade data available, standard proxies for trade costs, and product substitutability. In the last chapter, I analyze the empirical linkages between cartels and trade. First, I find that the average effect of cartels on trade is positive and significant. With regards to the effects of multi-product collusion on bilateral trade, the results show that the impact of multi-product cartels on trade is positive, significant and statistically larger than the effect of single-product cartels, consistent with my model. The positive and significant impact of multi-product cartels on trade becomes more pronounced when the goods are sufficiently unrelated, in line with the theoretical model. Moreover, I propose a two-stage estimation procedure to examine empirically 1) the relations between cartel discipline and (internal and external) trade costs; 2) the relation between (internal and external) trade and cartel discipline. As predicted by the theory, both internal trade costs and external trade costs are inversely related to collusive discipline. Using the first-stage estimates, I construct different measures of cartel discipline, and in the second-stage analysis I find the effect of cartel discipline on both internal and external trade to be negative and significant, in line with the theoretical predictions.

Metrics

38 File views/ downloads
24 Record Views

Details

Logo image