# An empirical analysis of the spatial variability of fuel prices in the United States

## TL;DR

This study analyzes the geographic variation of fuel prices across the US using high-resolution data, revealing socio-economic and local factors influencing prices, and introduces a theoretical model of station competition.

## Contribution

It provides a comprehensive empirical analysis of fuel price variability, combining statistical methods and a new theoretical model to understand local and regional influences.

## Key findings

- Significant influence of median income and wages on fuel prices.
- Identification of an optimal spatial range for stationarity.
- Population density impacts prices in line with a competition model.

## Abstract

In this paper, we use a newly constructed dataset to study the geographic distribution of fuel price across the US at a very high resolution. We study the influence of socio-economic variables through different and complementary statistical methods. We highlight an optimal spatial range roughly corresponding to stationarity scale, and significant influence of variables such as median income, wage with a non-simple spatial behavior that confirms the importance of geographical particularities. On the other hand, multi-level modeling reveals a strong influence of the state in the level of price but also of some local characteristics including population density. Through the combination of such methods, we unveil the superposition of a governance process with a local socio-economical spatial process. The influence of population density on prices is furthermore consistent with a minimal theoretical model of competition between gas stations, that we introduce and solve numerically. We discuss developments and applications, including the elaboration of locally parametrized car-regulation policies.