In oil-producing economies with fuel subsidies, the low cost of private vehicle use may weaken the conventional relationship between urban density and gasoline consumption. This study examines the relationship between urban population density and per capita gasoline consumption using a balanced panel of 80 Iranian counties, comprising 800 county-year observations over the 2013–2022 period. To this end, administrative data on recorded gasoline sales were combined with demographic, economic, and transportation data, and linear and nonlinear models were estimated using county and year fixed effects and county-clustered standard errors. Heterogeneity in this relationship across small, medium-sized, and large counties was also tested by interacting county-size categories with the linear and quadratic terms of population density. The linear model indicates that, after controlling for other factors, a 1% increase in population density is associated, on average, with an approximately 0.56% reduction in per capita gasoline consumption. However, the nonlinear model shows that the strength of this relationship is not constant and becomes more negative as density increases. Within the common density range, the estimated elasticities change from −0.832 to −0.949 for small counties, from −0.383 to −0.734 for medium-sized counties, and from −0.824 to −0.913 for large counties; all estimates are statistically significant. Overall, the findings suggest that compact development policies should be tailored to each county’s initial density level and spatial structure and coordinated with access to employment, services, and public transportation.