Purpose - The demand for alcohol is a well-researched topic, yet the published literature regarding consumer responses to changes in the price of alcohol includes many conflicting and inconsistent results. This paper seeks to present an analysis of known own-price elasticity estimates, and to attempt to understand why reported results differ.Design/methodology/approach - The approach taken to analysing the data is the meta-regression approach. Specifically, the meta-regression considers 150 beers, spirits, and wines, own price elasticity point estimates, which have been drawn from studies that consider demand responses to changes in the price of alcohol in 18 different countries.Findings - The results of the empirical work reported in this paper suggest that the year of the study, the length of study, the per capita level of alcohol consumption, and the relative ethanol share of a beverage are important factors when explaining variations in consumer demand responses to changes in the price of alcohol. Interestingly, the study also suggests that country-specific and beverage-specific effects are not important.Originality/value - The paper is valuable as it uses the meta-regression framework to control for study design characteristics and, once these characteristics are controlled for, it becomes possible to identify the underlying trend in the demand for alcohol. Specifically, the trend was shown to be one where the demand for alcoholic beverages became increasingly inelastic up to 1969 and decreasingly inelastic thereafter.