Stochastic Frontier Estimation of Market Power in the Food Industries

Abstract

We use a modified version of the stochastic frontier model to estimate oligopoly markups above the perfectly competitive frontier, separating out deterministic markups from purely stochastic markups. Using data from 42 US food processing industries between 1990 and 2010, empirical results indicate a widespread incidence of oligopoly power, with Lerner indexes averaging approximately 21%. Further, the estimated markups increase with industrial concentration but decrease with price elasticities and imports. Finally, the estimated Lerner indexes are in the range of previous food industry estimates using New Empirical Industrial Organization (NEIO) models.

A Stochastic Frontier Analysis Approach for Estimating Market Power in the Major US Meat Export Markets

“…Mean markup 0.744*** 0.011 ('***', '**', '*'): 1%, 5% and 10% level of significance, respectively. There are two main differences regarding the different empirical findings of the study by Lopez et al (2018) . Secondly, there are some distinct structural differences between the two poultry and the beef sector that can justify the different outcomes.…”

“…The SF method treats mark-ups as deviations from an optimal marginal cost pricing frontier. This methodology has been applied to the U.S. food industry by Lopez et al (2018), to the U.S. cattle industry by Panagiotou and Stavrakoudis (2017), and to the Brazilian milk market by Scalco et al (2017).…”

“…The goal of the present study is to examine the exertion of oligopoly power in international meat export markets. Panagiotou andStavrakoudis (2017, 2018) and Lopez et al (2018) have employed the stochastic frontier methodology in order to obtain market power estimates for the domestic meat industry. exerted by the U.S. pork packing industry was 4.317% and 3.530%, respectively.…”

“…One of the big advantages of the SF estimation approach is that it bypasses the estimation of demand and conduct needed in new empirical industrial organization (NEIO) models in order to measure the gap between price and marginal cost of production (Lopez et al, 2018). In the present study, where data for estimating demand for US meat exports by foreign countries would be hard to collect, the SF estimation technique is the most indicative in order to obtain market power estimates in each one of the meat export markets under examination.…”

“…Average estimates of the degree of market power (θ) and the Lerner index (L%)Lopez et al (2018) found that for the poultry processors for the period 1990-2011, the average Lerner index (LI) was around 17%. This result is similar to the Lerner index for the pork exports (16.5%) estimated in this work.…”