Essays on retail gasoline pricing
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
This dissertation investigates retailers' pricing behavior by focusing on the retail gasoline market in Seoul, South Korea. The study consists of two chapters. The first chapter presents an empirical analysis of how gasoline stations change prices. One common feature of retail price changes is their periodic and lumpy nature. Various theories have been proposed to explain why retail prices are sticky, each focusing on different factors that influence how retailers set prices. Studies in macroeconomic theory indicate that optimal price adjustment patterns depend on the nature of costs involved in price adjustment. If a price change incurs a fixed cost (e.g., a "menu cost"), optimal price changes occur when the change in state variables exceeds a threshold, resulting in state-dependent price changes. Alternatively, if there is a fixed cost associated with acquiring information, it is optimal to make price adjustments with periodic regularity (time-dependent). Studies in industrial organization emphasize the role of market power and strategic interactions among retailers. Some studies in marketing science argue that retailers are more likely to maintain certain prices based on consumer psychology (e.g., those ending with the digit 9). Do some or all of these theories explain how Seoul's gasoline retail price changes? In the first chapter, I examine the empirical significance of time-dependent pricing, state-dependent pricing, market power, and psychological pricing in the estimation model and examine how these factors are correlated with each other. The estimation results show that the most dominant factor affecting pricing decisions is the time dependent pricing rule, and this tendency to follow the time-dependent pricing rule varies with retailers' local market power. The analysis of how frequently gas stations change prices helps to better understand another common empirical phenomenon concerning asymmetric changes in retail gasoline prices in response to changes in wholesale prices. Many studies find that increases in costs (such as oil prices) are passed through more quickly to retail prices than decreases in costs, a pattern known as "rockets and feathers". This literature is mostly based on the error correction model that assumes retail prices are a linear function of costs. The pricing behavior examined in the first chapter implies that changes in retail prices are nonlinear in changes in costs, as gas stations keep daily prices mostly unchanged despite continuous changes in costs. In the second chapter, I demonstrate the potential bias arising from using daily data for the "rockets and feathers" study. Recent studies on "rockets and feathers" tend to rely more on high-frequency data to avoid bias arising from the temporal aggregation of data. In this study, I investigate price adjustment patterns by estimating an error correction model using daily station-level data from the Korean gasoline market. I find that compared to those based on weekly data, the estimated adjustment patterns based on daily data exhibit greater variation, which may be attributed to model misspecification that fails to account for the essential feature of daily-level data: censored responses to cost changes. The empirical findings emphasize the need for careful model specification when investigating the price adjustment pattern with daily-level data. In additional analyses, I explore the effect of consumer search on adjustment patterns and find that consumer search may not be a primary driving factor behind asymmetric price adjustments.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.