A Bayesian heterogeneous coefficients spatial autoregressive panel data model of retail fuel duopoly pricing

Abstract

We apply a heterogenous coefficient spatial autoregressive panel model to explore competition/cooperation by duopoly pairs of German fueling stations in setting prices for diesel and E5 fuel. We rely on a Markov Chain Monte Carlo (MCMC) estimation methodology applied with non-informative priors, which produces estimates equivalent to those from (quasi-) maximum likelihood. We explore station-level pricing behavior using pairs of proximately situated fueling stations with no nearby neighbors. Our sample data represents average daily diesel and e5 fuel prices, and refinery cost information covering more than 487 days.The heterogeneous coefficients spatial autoregressive panel data model uses the large sample of daily time periods to produce spatial autoregressive model estimates for each fueling station. These estimates provide information regarding the price reaction function of each station to its duopoly rival station. This is in contrast to conventional estimates of price reaction functions that average over the entire cross-sectional sample of stations. We show how these estimates can be used to infer competition versus cooperation in price setting by individual stations.

Markov Chain Monte Carlo estimation of spatial dynamic panel models for large samples

“…In this vein, LeSage et al (2017) describe a balanced panel of daily station-level prices created for over 14,000 filling stations in Germany, operating over the period from June 1, 2014 to September 30, 2015, or T = 487 days. Price for both diesel and e5 fuels are in nominal terms and include excise and value-added taxes.…”

“…While both Kihm et al (2016) and LeSage et al (2017) consider the degree of competition in the vicinity of the station, they do not further examine the competition/noncompetition behavior across different neighboring brands. Other things equal, proximity to stations with dominating brand name should lead to brand competition on price, while proximity to the non-major brand stations might be expected to reduce the tension of brand competition.…”

“…An illustration of the method uses a sample of T = 487 daily gas station prices for more than N = 12, 000 stations in Germany (described in LeSage et al (2017)), resulting in N × T greater than 6 million. Fixed effects in spatial panel data models have been extensively analyzed for the case of large T by Yu et al (2008), and for small T in Su and Yang (2015).…”

A Matrix Exponential Spatial Panel Model with Heterogeneous Coefficients

“…This could be interpreted as evidence in favor of tax competition between regions in our sample. LeSage, Vance, and Chih (2017) argue that a more ideal situation would allow inference regarding how each of the individual observational units (gas stations in their case) i51; . .…”

“…LeSage, Vance, and Chih () argue that a more ideal situation would allow inference regarding how each of the individual observational units (gas stations in their case) react to pricing actions taken by each station i 's neighboring stations. Some stations might exhibit competitive reactions vis‐à‐vis their neighbors, while others react in a cooperative fashion, or do not react at all.…”